<?xmlversion="1.0" encoding="US-ASCII"?>version='1.0' encoding='utf-8'?> <!DOCTYPE rfcSYSTEM "rfc2629.dtd"> <?rfc toc="yes"?> <?rfc tocompact="yes"?> <?rfc tocdepth="3"?> <?rfc tocindent="yes"?> <?rfc symrefs="yes"?> <?rfc sortrefs="yes"?> <?rfc comments="yes"?> <?rfc inline="yes"?> <?rfc compact="yes"?> <?rfc subcompact="no"?>[ <!ENTITY nbsp " "> <!ENTITY zwsp "​"> <!ENTITY nbhy "‑"> <!ENTITY wj "⁠"> ]> <rfc xmlns:xi="http://www.w3.org/2001/XInclude" category="std" docName="draft-ietf-ippm-initial-registry-16"ipr="trust200902">number="8912" ipr="trust200902" obsoletes="" updates="" submissionType="IETF" consensus="true" xml:lang="en" tocInclude="true" tocDepth="3" symRefs="true" sortRefs="true" version="3"> <!-- xml2rfc v2v3 conversion 2.43.0 --> <front> <title abbrev="Initial Performance Metrics Registry">Initial Performance Metrics Registry Entries</title> <seriesInfo name="RFC" value="8912"/> <author fullname="Al Morton" initials="A." surname="Morton"> <organization>AT&T Labs</organization> <address> <postal> <street>200 Laurel Avenue South</street><city>Middletown,</city><city>Middletown</city> <region>NJ</region> <code>07748</code><country>USA</country><country>United States of America</country> </postal> <phone>+1 732 420 1571</phone><facsimile>+1 732 368 1192</facsimile><email>acmorton@att.com</email><uri/></address> </author> <author fullname="Marcelo Bagnulo" initials="M." surname="Bagnulo"> <organization abbrev="UC3M">Universidad Carlos III de Madrid</organization> <address> <postal> <street>Av. Universidad 30</street> <city>Leganes</city> <region>Madrid</region> <code>28911</code><country>SPAIN</country><country>Spain</country> </postal> <phone>34 91 6249500</phone> <email>marcelo@it.uc3m.es</email> <uri>http://www.it.uc3m.es</uri> </address> </author> <author fullname="Philip Eardley" initials="P." surname="Eardley"> <organization abbrev="BT">BT</organization> <address> <postal> <street>Adastral Park, Martlesham Heath</street> <city>Ipswich</city><country>ENGLAND</country><country>United Kingdom</country> </postal> <email>philip.eardley@bt.com</email> </address> </author> <author fullname="Kevin D'Souza" initials="K." surname="D'Souza"> <organization>AT&T Labs</organization> <address> <postal> <street>200 Laurel Avenue South</street><city>Middletown,</city><city>Middletown</city> <region>NJ</region> <code>07748</code><country>USA</country><country>United States of America</country> </postal> <phone>+1 732 420xxxx</phone> <facsimile/>2514</phone> <email>kld@att.com</email><uri/></address> </author> <dateday="9" month="March" year="2020"/>month="November" year="2021"/> <keyword>Loss</keyword> <keyword>Delay</keyword> <keyword>Delay Variation</keyword> <keyword>ICMP ping</keyword> <keyword>DNS Response</keyword> <keyword>Poisson</keyword> <keyword>Periodic</keyword> <keyword>TCP</keyword> <abstract> <t>This memo defines the set ofInitial Entriesinitial entries for the IANA Registry of PerformanceMetrics Registry.Metrics. The setincludes:includes UDPRound-tripRound-Trip Latency and Loss, Packet Delay Variation, DNS Response Latency and Loss, UDP PoissonOne-wayOne-Way Delay and Loss, UDP PeriodicOne-wayOne-Way Delay and Loss, ICMPRound-tripRound-Trip Latency and Loss, and TCPround-trip LatencyRound-Trip Delay and Loss.</t> </abstract><note title="Requirements Language"> <t>The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14<xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they appear in all capitals, as shown here.</t> <t/> </note></front> <middle> <sectiontitle="Introduction">numbered="true" toc="default"> <name>Introduction</name> <t>This memoproposesdefines an initial set of entries for the Performance Metrics Registry. It uses terms and definitions from theIPPMIP Performance Metrics (IPPM) literature, primarily <xreftarget="RFC2330"/>.</t>target="RFC2330" format="default"/>.</t> <t>Although there are several standard templates for organizing specifications ofperformance metricsPerformance Metrics (see <xreftarget="RFC7679"/>target="RFC7679" format="default"/> for an example of the traditional IPPM template, based to a large extent on the Benchmarking Methodology Working Group's traditional template in <xreftarget="RFC1242"/>,target="RFC1242" format="default"/>, and see <xreftarget="RFC6390"/>target="RFC6390" format="default"/> for a similar template), none of these templates were intended to become the basis for the columns of an IETF-wideregistryRegistry of metrics. While examining aspects of metric specificationswhichthat need to be registered, it became clear that none of the existing metric templates fullysatisfiessatisfy the particular needs of aregistry.</t>Registry.</t> <t>Therefore, <xreftarget="I-D.ietf-ippm-metric-registry"/>target="RFC8911" format="default"/> defines the overall format for a Performance Metrics Registry.Section 5 of<xreftarget="I-D.ietf-ippm-metric-registry"/>target="RFC8911" sectionFormat="of" section="5"/> also gives guidelines for those requesting registration of aMetric,Metric -- thatisis, the creation ofentry(s)one or more entries in the Performance MetricsRegistry: "InRegistry:</t> <blockquote>In essence, there needs to be evidence that (1) a candidate Registered Performance Metric has significant industryinterest,interest or has seendeployment,deployment and (2) there is agreement that the candidate Registered Performance Metric serves its intendedpurpose." Thepurpose.</blockquote> <t>The process defined in <xreftarget="I-D.ietf-ippm-metric-registry"/>target="RFC8911" format="default"/> also requires that new entriesarebe administered by IANA through the Specification Requiredpolicy,policy <xref target="RFC8126"/>, which will ensure that the metrics are tightly defined.</t> <section numbered="true" toc="default"> <name>Requirements Language</name> <t>The key words "<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>", "<bcp14>MAY</bcp14>", and "<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as described in BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they appear in all capitals, as shown here.</t> </section> </section> <sectiontitle="Scope">numbered="true" toc="default"> <name>Scope</name> <t>This document defines a set of initial Performance Metrics Registryentries.Entries. Most are Active Performance Metrics, which are based on RFCs prepared in the IPPMworking groupWorking Group of the IETF, according to their framework <xreftarget="RFC2330"/>target="RFC2330" format="default"/> and its updates.</t> </section> <!-- Section 3--> <sectiontitle="Registrynumbered="true" toc="default"> <name>Registry Categories andColumns">Columns</name> <t>This memo uses the terminology defined in <xreftarget="I-D.ietf-ippm-metric-registry"/>.</t>target="RFC8911" format="default"/>.</t> <t>This section provides the categories and columns of theregistry,Registry, for easy reference. An entry (row) therefore gives a complete description of a Registered Metric.</t><t><figure> <artwork><![CDATA[Legend: Registry<t>Registry Categories andColumns,Columns are shownasbelow in this format:</t> <artwork name="" type="" align="left" alt=""><![CDATA[ Category------------------------------------... Column | Column||... ]]></artwork> <artwork name="" type="" align="left" alt=""><![CDATA[ Summary--------------------------------------------------------------------------------------------------------------------------------------- Identifier | Name | URI | Desc. | Reference | ChangeController| Ver | | | | | | Controller | Metric Definition ----------------------------------------- Reference Definition | Fixed Parameters | Method of Measurement --------------------------------------------------------------------- Reference | Packet | Traffic | Sampling |Run-timeRuntime | Role | Method | Stream | Filter | Distribution | Parameters | | | Generation | Output ----------------------------------------- Type | Reference | Units | Calibration | | Definition | | | Administrative Information------------------------------------------------------------------------- Status |Requester | Rev |Rev.DateRev. Date | Comments and Remarks -------------------- ]]></artwork></figure></t></section> <!-- Section 4 --> <sectiontitle="UDP Round-tripanchor="udp-rt-latency-loss-reg-entries" numbered="true" toc="default"> <name>UDP Round-Trip Latency and Loss RegistryEntries">Entries</name> <t>This section specifies an initialregistry entryRegistry Entry fortheUDPRound-trip Latency,Round-Trip Latency and another entry for the UDPRound-tripRound-Trip Loss Ratio.</t><t>Note:<t indent="3">Note: Each RegistryentryEntry only produces a "raw" output or a statistical summary. To describe both "raw" and one or more statistics efficiently, the Identifier, Name, and OutputCategoriescategories can besplitsplit, and a single section can specify two or moreclosely-relatedclosely related metrics. For example, this section specifies two RegistryentriesEntries with many common columns. SeeSection 7<xref target="udp-poisson-owd-owl-reg"/> for an example specifying multiple RegistryentriesEntries with many common columns.</t> <t>All column entriesbesidebesides the ID, Name, Description, and Output Reference Method categories are thesame, thussame; thus, this sectionproposesdefines twoclosely-related registry entries.closely related Registry Entries. As a result, IANAishas alsoasked to assignassigned a corresponding URL to each of the two NamedMetric.</t>Metrics.</t> <!-- 4.1 --> <sectiontitle="Summary">numbered="true" toc="default"> <name>Summary</name> <t>This category includes multiple indexes to theregistry entry:Registry Entries: the element ID andmetric name.</t>Metric Name.</t> <!-- 4.1.1 --> <sectiontitle="ID (Identifier)">numbered="true" toc="default"> <name>ID (Identifier)</name> <t>IANAis asked to assign differenthas allocated the numericidentifiers to each ofIdentifiers 1 and 2 for the two NamedMetrics.</t>Metric Entries in <xref target="udp-rt-latency-loss-reg-entries"/>. See <xref target="name412"/> for mapping to Names. </t> </section> <!-- 4.1.2 --> <sectiontitle="Name"> <t>RTDelay_Active_IP-UDP-Periodic_RFCXXXXsec4_Seconds_95Percentile</t> <t>RTLoss_Active_IP-UDP-Periodic_RFCXXXXsec4_Percent_LossRatio</t>anchor="name412" numbered="true" toc="default"> <name>Name</name> <dl> <dt>1:</dt><dd>RTDelay_Active_IP-UDP-Periodic_RFC8912sec4_Seconds_95Percentile</dd> <dt>2:</dt><dd>RTLoss_Active_IP-UDP-Periodic_RFC8912sec4_Percent_LossRatio</dd> </dl> </section> <!-- 4.1.3 --> <sectiontitle="URI">numbered="true" toc="default"> <name>URI</name> <t>URL:https://www.iana.org/ ... <name></t><eref target="https://www.iana.org/performance-metrics/RTDelay_Active_IP-UDP-Periodic_RFC8912sec4_Seconds_95Percentile"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTLoss_Active_IP-UDP-Periodic_RFC8912sec4_Percent_LossRatio"/></t> </section> <!-- 4.1.4 --> <sectiontitle="Description"> <t>RTDelay: Thisnumbered="true" toc="default"> <name>Description</name> <dl newline="false" spacing="normal"> <dt>RTDelay:</dt> <dd>This metric assesses the delay of a stream of packets exchanged between two hosts (which are the two measurementpoints), and the Outputpoints). The output is theRound-tripround-trip delay for all successfully exchanged packets expressed as the 95th percentile of their conditional delaydistribution.</t> <t>RTLoss: Thisdistribution.</dd> <dt>RTLoss:</dt> <dd>This metric assesses the loss ratio of a stream of packets exchanged between two hosts (which are the two measurementpoints), and the Outputpoints). The output is theRound-tripround-trip loss ratio for allsuccessfully exchangedtransmitted packets expressed as apercentage.</t>percentage.</dd> </dl> </section> <!-- 4.1.5 --> <sectiontitle="Change Controller">numbered="true" toc="default"> <name>Change Controller</name> <t>IETF</t> </section> <!-- 4.1.6 --> <sectiontitle="Versionnumbered="true" toc="default"> <name>Version (of RegistryFormat)">Format)</name> <t>1.0</t> </section> </section> <!-- 4.2 --> <sectiontitle="Metric Definition">numbered="true" toc="default"> <name>Metric Definition</name> <t>This category includes columns to prompt the entry of all necessary details related to the metric definition, including the RFC reference and values of input factors, calledfixed parameters.</t>"Fixed Parameters".</t> <!-- 4.2.1 --> <sectiontitle="Reference Definition"> <t>Almes,numbered="true" toc="default"> <name>Reference Definition</name> <t>For delay:</t> <t indent="3">Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip Delay Metric for IPPM", RFC 2681, DOI 10.17487/RFC2681, September1999.</t> <t><xref target="RFC2681"/></t> <t>Section 2.4 of1999, <https://www.rfc-editor.org/info/rfc2681>. <xreftarget="RFC2681"/>target="RFC2681"/></t> <t indent="3"><xref target="RFC2681" sectionFormat="of" section="2.4"/> provides the reference definition of the singleton (single value)Round-tripround-trip delay metric.Section 3.4 of<xreftarget="RFC2681"/>target="RFC2681" sectionFormat="of" section="3.4"/> provides the reference definition expanded to cover a multi-singleton sample. Note that terms such assingleton"singleton" andsample"sample" are defined inSection 11 of<xreftarget="RFC2330"/>.</t> <t>Notetarget="RFC2330" sectionFormat="of" section="11"/>.</t> <t indent="3">Note that although the<xref target="RFC2681"/>definition of"Round-trip-Delayround-trip delay betweenSrcthe Source (Src) andDst"the Destination (Dst) as provided in <xref target="RFC2681" sectionFormat="of" section="2.4"/> is directionally ambiguous in the text, this metric tightens the definition further to recognize that the host in the"Src" roleSrc Role will send the first packet to"Dst",the host in the Dst Role and will ultimately receive the corresponding return packet from"Dst"the Dst (when neitherareis lost).</t><t>Finally,<t indent="3">Finally, note that the variable "dT" is used in <xreftarget="RFC2681"/>target="RFC2681" format="default"/> to refer to the value ofRound-tripround-trip delay in metric definitions and methods. The variable "dT" has beenre-usedreused in other IPPM literature to refer to differentquantities,quantities and cannot be used as a global variable name.</t><t>Morton,<t>For loss:</t> <t indent="3">Morton, A.,"Round-trip"Round-Trip Packet Loss Metrics", RFC 6673, DOI 10.17487/RFC6673, August2012.</t> <t><xref2012, <https://www.rfc-editor.org/info/rfc6673>. <xref target="RFC6673"/></t> <t>BothdelayDelay andlossLoss metrics employ a maximum waiting time for received packets, so the count of lost packets to total packets sent is the basis for the loss ratio calculation as perSection 6.1 of<xreftarget="RFC6673"/>.</t>target="RFC6673" sectionFormat="of" section="6.1"/>.</t> </section> <!-- 4.2.2 --> <sectiontitle="Fixed Parameters"> <t>Type-Pnumbered="true" toc="default"> <name>Fixed Parameters</name> <dl newline="true" spacing="normal"> <dt>Type-P as defined inSection 13 of<xreftarget="RFC2330"/>: <list style="symbols"> <t>IPv4target="RFC2330" sectionFormat="of" section="13"/>:</dt><dd><t/> <dl newline="true" spacing="normal"> <dt>IPv4 headervalues: <list style="symbols"> <t>DSCP: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to0</t> <t>TTL: set0</dd> <dt>TTL:</dt><dd>Set to255</t> <t>Protocol: set255</dd> <dt>Protocol:</dt><dd>Set to 17(UDP)</t> </list></t> <t>IPv6(UDP)</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>IPv6 headervalues:<list style="symbols"> <t>DSCP: set to 0</t> <t>Hop Count: set to 255</t> <t>Next Header: setvalues:</dt> <dd><t/><dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to 0</dd> <dt>Hop Count:</dt><dd>Set to 255</dd> <dt>Next Header:</dt><dd>Set to 17(UDP)</t> <t>Flow Label: set to zero</t> <t>Extension Headers: none</t> </list></t> <t>UDP(UDP)</dd> <dt>Flow Label:</dt><dd>Set to 0</dd> <dt>Extension Headers:</dt><dd>None</dd> </dl></dd> </dl> <dl newline="true" spacing="normal"> <dt>UDP headervalues: <list style="symbols"> <t>Checksum: thevalues:</dt> <dd><t/><dl newline="false" spacing="compact"> <dt>Checksum:</dt><dd>The checksumMUST<bcp14>MUST</bcp14> be calculated and the non-zero checksum included in theheader</t> </list></t> <t>UDP Payload <list style="symbols"> <t>totalheader</dd> </dl></dd> </dl> <dl newline="true" spacing="normal"> <dt>UDP Payload:</dt> <dd><t/><dl newline="false" spacing="compact"> <dt>Total of 100bytes</t> </list></t> </list></t> <t>Other measurement parameters:<list style="symbols"> <t>Tmax: abytes</dt><dd/> </dl></dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>Other measurement Parameters:</dt> <dd><t/> <dl newline="false" spacing="normal"> <dt>Tmax:</dt><dd>A loss threshold waitingtime<list style="symbols"> <t>3.0,time with value 3.0, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1 ms), with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> </list></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </dd> </dl> </section> </section> <!-- 4.3 --> <sectiontitle="Methodnumbered="true" toc="default"> <name>Method ofMeasurement">Measurement</name> <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguousmethodsmethod for implementations.</t> <!-- 4.3.1 --> <sectiontitle="Reference Method">numbered="true" toc="default"> <name>Reference Methods</name> <t>The methodology for this metric (equivalent to Type-P-Round-trip- Delay and Type-P-Round-trip-Delay-Poisson-Stream) is defined asType-P-Round-trip-Delay-Poisson-Streaminsection 2.6 of<xreftarget="RFC2681">RFC 2681</xref>target="RFC2681" sectionFormat="of" section="2.6"/> (for singletons) andsection 3.6 of<xreftarget="RFC2681">RFC 2681</xref>target="RFC2681" sectionFormat="of" section="3.6"/> (for samples) using the Type-P and Tmax definedunderin the FixedParameters.Parameters column. However, the Periodic stream will be generated according to <xreftarget="RFC3432"/>.</t>target="RFC3432" format="default"/>.</t> <t>The reference method distinguishes between long-delayed packets and lost packets by implementing a maximum waiting time for packet arrival. Tmax is the waiting time used as the threshold to declare a packet lost. Lost packetsSHALL<bcp14>SHALL</bcp14> be designated as having undefineddelay,delay and counted for the RTLossmetric.</t>metric <xref target="RFC6673"/>.</t> <t>The calculations on the delay (RTT)SHALL<bcp14>SHALL</bcp14> be performed on the conditional distribution, conditioned on successful packet arrival within Tmax. Also, when all packet delays are stored, the processwhichthat calculates the RTT valueMUST<bcp14>MUST</bcp14> enforce the Tmax threshold on stored values before calculations. Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>The reference method requires some way to distinguish between different packets in a stream to establish correspondence between sending times and receiving times for eachsuccessfully-arrivingsuccessfully arriving packet. Sequence numbers or other send-order identificationMUST<bcp14>MUST</bcp14> be retained at the Src or included with each packet to disambiguate packet reordering if it occurs.</t> <t>If a standard measurement protocol is employed, then the measurement process will determine the sequence numbers or timestamps applied to test packets after the Fixed and RuntimeparametersParameters are passed to that process. The chosen measurement protocol will dictate the format of sequence numbers andtime-stamps,timestamps, if they are conveyed in the packet payload.</t> <t>Refer toSection 4.4 of<xreftarget="RFC6673"/>target="RFC6673" sectionFormat="of" section="4.4"/> for an expanded discussion of the instruction to "send a Type-P packet back to the Src as quickly as possible" inSection 2.6 of <xref target="RFC2681">RFC 2681</xref>. Section 8 of<xreftarget="RFC6673"/>target="RFC2681" sectionFormat="of" section="2.6"/>. <xref target="RFC6673" sectionFormat="of" section="8"/> presents additional requirementswhich MUSTthat <bcp14>MUST</bcp14> be included in themethodMethod ofmeasurementMeasurement for this metric.</t> </section> <!-- 4.3.2 --> <sectiontitle="Packetnumbered="true" toc="default"> <name>Packet StreamGeneration">Generation</name> <t>This sectiongives theprovides detailsof theregarding packettraffictraffic, which is used as the basis for measurement. In IPPMmetrics,Metrics, this is called theStream, and"stream"; this stream can easily be described by providing the list of streamparameters.</t> <t>Section 3 of <xref target="RFC3432"/>Parameters.</t> <t><xref target="RFC3432" sectionFormat="of" section="3"/> prescribes the method for generating Periodic streams using associatedparameters.</t> <t><list style="hanging"> <t hangText="incT">theParameters.</t> <dl newline="false" spacing="normal"> <dt>incT:</dt> <dd>The nominal duration of the inter-packet interval, first bit to first bit, with value 0.0200, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1ms).</t> <t hangText="dT">thems).</dd> <dt>dT:</dt> <dd>The duration of the interval for allowed sample start times, with value 1.0, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1ms).</t> </list>NOTE: anms).</dd> </dl> <t indent="3">Note: An initiation process with a number of control exchanges resulting in unpredictable start times (within a time interval) may be sufficient to avoid synchronization of periodicstreams,streams andthereforeis a valid replacement for selecting a start time at random from a fixed interval.</t> <t>The T0parameterParameter will be reported as a measuredparameter.Parameter. Parameters incT and dT are Fixed Parameters.</t> </section> <!-- 4.3.3 --> <sectiontitle="Trafficnumbered="true" toc="default"> <name>Traffic Filtering(observation) Details"> <t>NA</t>(Observation) Details</name> <t>N/A</t> </section> <!-- 4.3.4 --> <sectiontitle="Sampling Distribution"> <t>NA</t>numbered="true" toc="default"> <name>Sampling Distribution</name> <t>N/A</t> </section> <!-- 4.3.5 --> <sectiontitle="Run-timenumbered="true" toc="default"> <name>Runtime Parameters and DataFormat"> <t>Run-timeFormat</name> <t>Runtime Parameters are input factors that must be determined, configured into the measurement system, and reported with the results for the context to be complete.</t><t><list style="hanging"> <t hangText="Src">the<dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>The IP address of the host in the Src Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seeSection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="Dst">thetarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>Dst:</dt> <dd>The IP address of the host in the Dst Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seesection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="T0">atarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>T0:</dt> <dd>A time, the start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros", a start time is unspecified and Tf is to be interpreted as theDurationduration of the measurement interval. The start time is controlled through othermeans.</t> <t hangText="Tf">ameans.</dd> <dt>Tf:</dt> <dd>A time, the end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros",a endan ending time and date is ignored and Tf is interpreted as theDurationduration of the measurementinterval.</t> </list></t> <t/>interval.</dd> </dl> </section> <!-- 4.3.6 --> <sectiontitle="Roles"> <t><list style="hanging"> <t hangText="Src">launchesnumbered="true" toc="default"> <name>Roles</name> <dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>Launches each packet and waits for return transmissions fromDst.</t> <t hangText="Dst">waitsthe Dst.</dd> <dt>Dst:</dt> <dd>Waits for each packet from the Src and sends a return packet toSrc.</t> </list></t>the Src.</dd> </dl> </section> </section> <!-- 4.4 --> <sectiontitle="Output">numbered="true" toc="default"> <name>Output</name> <t>This category specifies all details of theOutputoutput of measurements using the metric.</t> <!-- 4.4.1 --> <sectiontitle="Type"> <t>Percentile -- fornumbered="true" toc="default"> <name>Type</name> <t>Percentile: For the conditional distribution of all packets with a valid value ofRound-tripround-trip delay (undefined delays are excluded), this is a single value corresponding to the 95th percentile, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>The percentile = 95, meaning that the reported delay, "95Percentile", is the smallest value ofRound-tripround-trip delay for which the Empirical DistributionFunction (EDF), F(95Percentile) >=Function, EDF(95Percentile), is greater than or equal to 95% of the singletonRound-tripround-trip delay values in the conditional distribution. Seesection 11.3 of<xreftarget="RFC2330"/>target="RFC2330" sectionFormat="of" section="11.3"/> for the definition of the percentile statistic using the EDF.</t><t>LossRatio --<t>For LossRatio, the count of lost packets to total packets sent is the basis for the loss ratio calculation as perSection 6.1 of<xreftarget="RFC6673"/>.</t>target="RFC6673" sectionFormat="of" section="6.1"/>.</t> </section> <!-- 4.4.2 --> <sectiontitle="Reference Definition">numbered="true" toc="default"> <name>Reference Definition</name> <t>For alloutputs ---</t> <t><list style="hanging"> <t hangText="T0">theoutputs:</t> <dl newline="false" spacing="normal"> <dt>T0:</dt> <dd>The start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="Tf">thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>Tf:</dt> <dd>The end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="TotalPkts">thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>TotalPkts:</dt> <dd>The count of packets sent by the Src to the Dst during the measurementinterval.</t> </list></t> <t/> <t>For</t> <t>RTDelay_Active_IP-UDP-Periodic_RFCXXXXsec4_Seconds_95Percentile:</t> <t><list style="hanging"> <t hangText="95Percentile">Theinterval.</dd> </dl> <dl newline="false" spacing="normal"> <dt>95Percentile:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0ns).</t> </list></t> <t>For</t> <t>RTLoss_Active_IP-UDP-Periodic_RFCXXXXsec4_Percent_LossRatio:</t> <t><list style="hanging"> <t hangText="Percentile">Thens).</dd> </dl> <dl newline="false" spacing="normal"> <dt>Percent_LossRatio:</dt> <dd>The numeric value of the result is expressed in units of lost packets to total packets times 100%, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.0000000001.</t> </list></t>0.0000000001.</dd> </dl> </section> <!-- 4.4.3 --> <sectiontitle="Metric Units">numbered="true" toc="default"> <name>Metric Units</name> <t>The 95thPercentilepercentile ofRound-trip Delayround-trip delay is expressed in seconds.</t> <t>TheRound-trip Loss Ratioround-trip loss ratio is expressed as a percentage of lost packets to total packets sent.</t> </section> <!-- 4.4.4 --> <sectiontitle="Calibration"> <t>Section 3.7.3 of <xref target="RFC7679"/>numbered="true" toc="default"> <name>Calibration</name> <t><xref target="RFC7679" sectionFormat="of" section="3.7.3"/> provides a means to quantify the systematic and random errors of a time measurement.In-situ calibrationCalibration in-situ could be enabled with an internal loopback at the Source host that includes as much of the measurement system as possible, performs address manipulation as needed, and provides some form of isolation (e.g., deterministic delay) to avoid send-receive interface contention. Some portion of the random and systematic error can be characterized in this way.</t> <t>When a measurement controller requests a calibration measurement, the loopback is applied and the result is output in the same format as a normalmeasurementmeasurement, with an additional indication that it is a calibration result.</t> <t>Both internal loopback calibration and clock synchronization can be used to estimate the available accuracy of the Output Metric Units. For example, repeated loopback delay measurements will reveal the portion of theOutputoutput result resolutionwhichthat is the result of systemnoise,noise and is thus inaccurate.</t> </section> </section> <!-- 4.5 --> <sectiontitle="Administrative items"> <t/>numbered="true" toc="default"> <name>Administrative Items</name> <!-- 4.5.1 --> <sectiontitle="Status">numbered="true" toc="default"> <name>Status</name> <t>Current</t> </section> <!-- 4.5.2 --> <sectiontitle="Requester"> <t>This RFC number</t>numbered="true" toc="default"> <name>Requester</name> <t>RFC 8912</t> </section> <!-- 4.5.3 --> <sectiontitle="Revision">numbered="true" toc="default"> <name>Revision</name> <t>1.0</t> </section> <!-- 4.5.4 --> <sectiontitle="Revision Date"> <t>YYYY-MM-DD</t>numbered="true" toc="default"> <name>Revision Date</name> <t>2021-11-17</t> </section> </section> <sectiontitle="Commentsnumbered="true" toc="default"> <name>Comments andRemarks"> <t>None.</t>Remarks</name> <t>None</t> </section> </section> <!-- Section 5 --> <sectiontitle="Packetanchor="packet_delay_variation" numbered="true" toc="default"> <name>Packet Delay Variation RegistryEntry">Entry</name> <t>This section gives an initialregistry entryRegistry Entry for a Packet Delay Variation (PDV) metric.</t> <sectiontitle="Summary">numbered="true" toc="default"> <name>Summary</name> <t>This category includes multiple indexes to theregistry entries,Registry Entry: the element ID andmetric name.</t>Metric Name.</t> <sectiontitle="ID (Identifier)"> <t><insertnumbered="true" toc="default"> <name>ID (Identifier)</name> <t>IANA has allocated the numericidentifier, an integer></t>Identifier 3 for the Named Metric Entry in <xref target="packet_delay_variation"/>. See <xref target="name512"/> for mapping to Name.</t> </section> <sectiontitle="Name"> <t>OWPDV_Active_IP-UDP-Periodic_RFCXXXXsec5_Seconds_95Percentile</t>anchor="name512" numbered="true" toc="default"> <name>Name</name> <dl> <dt>3:</dt><dd>OWPDV_Active_IP-UDP-Periodic_RFC8912sec5_Seconds_95Percentile</dd> </dl> </section> <sectiontitle="URI">numbered="true" toc="default"> <name>URI</name> <t>URL:https://www.iana.org/ ... <name></t><eref target="https://www.iana.org/performance-metrics/OWPDV_Active_IP-UDP-Periodic_RFC8912sec5_Seconds_95Percentile"/></t> </section> <sectiontitle="Description"> <t>An assessment ofnumbered="true" toc="default"> <name>Description</name> <t>This metric assesses packet delay variation with respect to the minimum delay observed on the periodicstream, and the Outputstream. The output is expressed as the 95th percentile of the one-way packet delay variation distribution.</t> </section> <sectiontitle="Change Controller">numbered="true" toc="default"> <name>Change Controller</name> <t>IETF</t> </section> <sectiontitle="Versionnumbered="true" toc="default"> <name>Version (of RegistryFormat)">Format)</name> <t>1.0</t> </section> </section> <sectiontitle="Metric Definition">numbered="true" toc="default"> <name>Metric Definition</name> <t>This category includes columns to prompt the entry of all necessary details related to the metric definition, including the RFC reference and values of input factors, calledfixed parameters.</t>"Fixed Parameters".</t> <sectiontitle="Reference Definition">numbered="true" toc="default"> <name>Reference Definition</name> <t>Paxson, V., Almes, G., Mahdavi, J., and M. Mathis, "Framework for IP Performance Metrics", RFC 2330, DOI 10.17487/RFC2330, May1998.1998, <https://www.rfc-editor.org/info/rfc2330>. <xref target="RFC2330"/></t> <t>Demichelis, C. and P. Chimento, "IP Packet Delay Variation Metric for IP Performance Metrics (IPPM)", RFC 3393, DOI 10.17487/RFC3393, November2002.2002, <https://www.rfc-editor.org/info/rfc3393>. <xref target="RFC3393"/></t> <t>Morton, A. and B. Claise, "Packet Delay Variation Applicability Statement", RFC 5481, DOI 10.17487/RFC5481, March2009.2009, <https://www.rfc-editor.org/info/rfc5481>. <xref target="RFC5481"/></t> <t>Mills, D., Martin, J., Ed., Burbank, J., and W. Kasch, "Network Time Protocol Version 4: Protocol and Algorithms Specification", RFC 5905, DOI 10.17487/RFC5905, June2010.<xref target="RFC5905"> </xref></t>2010, <https://www.rfc-editor.org/info/rfc5905>. <xref target="RFC5905"/></t> <t>Seesections 2.4Sections <xref target="RFC3393" section="2.4" sectionFormat="bare"/> and3.4<xref target="RFC3393" section="3.4" sectionFormat="bare"/> of <xref target="RFC3393"/>.SingletonThe measured singleton delay differencesmeasuredare referred to by the variable name "ddT" (applicable to all forms of delay variation). However, thismetric entryMetric Entry specifies the PDV form defined insection 4.2 of<xreftarget="RFC5481"/>,target="RFC5481" sectionFormat="of" section="4.2"/>, where the singleton PDV for packet i is referred to by the variable name "PDV(i)".</t> </section> <sectiontitle="Fixed Parameters"> <t><list style="symbols"> <t>IPv4numbered="true" toc="default"> <name>Fixed Parameters</name> <dl newline="true" spacing="normal"> <dt>IPv4 headervalues: <list style="symbols"> <t>DSCP: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to0</t> <t>TTL: set0</dd> <dt>TTL:</dt><dd>Set to255</t> <t>Protocol: set255</dd> <dt>Protocol:</dt><dd>Set to 17(UDP)</t> </list></t> <t>IPv6(UDP)</dd> </dl> </dd> <dt>IPv6 headervalues:<list style="symbols"> <t>DSCP: set to 0</t> <t>Hop Count: set to 255</t> <t>Next Header: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to 0</dd> <dt>Hop Count:</dt><dd>Set to 255</dd> <dt>Next Header:</dt><dd>Set to 17(UDP)</t> <t>Flow Label: set to zero</t> <t>Extension Headers: none</t> </list></t> <t>UDP(UDP)</dd> <dt>Flow Label:</dt><dd>Set to 0</dd> <dt>Extension Headers:</dt><dd>None</dd> </dl> </dd> <dt>UDP headervalues: <list style="symbols"> <t>Checksum: thevalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>Checksum:</dt><dd>The checksumMUST<bcp14>MUST</bcp14> be calculated and the non-zero checksum included in theheader</t> </list></t> <t>UDP Payload <list style="symbols"> <t>totalheader</dd> </dl> </dd> <dt>UDP Payload:</dt> <dd><t/><dl newline="false" spacing="compact"> <dt>Total of 200bytes</t> </list></t> </list></t> <t>Other measurement parameters:</t> <t><list style="hanging"> <t hangText="Tmax:">abytes</dt><dd/> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>Other measurement Parameters:</dt> <dd><t/> <dl newline="false" spacing="normal"> <dt>Tmax:</dt> <dd>A loss threshold waiting time with value 3.0, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1 ms), with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> <t hangText="F">atarget="RFC5905" sectionFormat="of" section="6"/>.</dd> <dt>F:</dt> <dd>A selection function unambiguously defining the packets from the stream selected for the metric. Seesection 4.2 of<xreftarget="RFC5481"/>target="RFC5481" sectionFormat="of" section="4.2"/> for the PDVform.</t> </list>Seeform.</dd> </dl> </dd> </dl> <t>See the Packet Streamgeneration categoryGeneration section for two additional Fixed Parameters.</t> </section> </section> <sectiontitle="Methodnumbered="true" toc="default"> <name>Method ofMeasurement">Measurement</name> <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguousmethodsmethod for implementations.</t> <sectiontitle="Reference Method">numbered="true" toc="default"> <name>Reference Methods</name> <t>Seesection 2.6Sections <xref target="RFC3393" section="2.6" sectionFormat="bare"/> and3.6<xref target="RFC3393" section="3.6" sectionFormat="bare"/> of <xref target="RFC3393"/> for general singleton element calculations. Thismetric entryMetric Entry requires implementation of the PDV form defined insection 4.2 of<xreftarget="RFC5481"/>.target="RFC5481" sectionFormat="of" section="4.2"/>. Also see measurement considerations insection 8 of<xreftarget="RFC5481"/>.</t>target="RFC5481" sectionFormat="of" section="8"/>.</t> <t>The reference method distinguishes between long-delayed packets and lost packets by implementing a maximum waiting time for packet arrival. Tmax is the waiting time used as the threshold to declare a packet lost. Lost packetsSHALL<bcp14>SHALL</bcp14> be designated as having undefined delay.</t> <t>The calculations on the one-way delaySHALL<bcp14>SHALL</bcp14> be performed on the conditional distribution, conditioned on successful packet arrival within Tmax. Also, when all packet delays are stored, the processwhichthat calculates the one-way delay valueMUST<bcp14>MUST</bcp14> enforce the Tmax threshold on stored values before calculations. Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>The reference method requires some way to distinguish between different packets in a stream to establish correspondence between sending times and receiving times for eachsuccessfully-arrivingsuccessfully arriving packet. Sequence numbers or other send-order identificationMUST<bcp14>MUST</bcp14> be retained at the Src or included with each packet to disambiguate packet reordering if it occurs.</t> <t>If a standard measurement protocol is employed, then the measurement process will determine the sequence numbers or timestamps applied to test packets after the Fixed and RuntimeparametersParameters are passed to that process. The chosen measurement protocol will dictate the format of sequence numbers andtime-stamps,timestamps, if they are conveyed in the packet payload.</t> </section> <sectiontitle="Packetnumbered="true" toc="default"> <name>Packet StreamGeneration">Generation</name> <t>This sectiongives theprovides detailsof theregarding packettraffictraffic, which is used as the basis for measurement. In IPPMmetrics,Metrics, this is called theStream, and"stream"; this stream can easily be described by providing the list of streamparameters.</t> <t>Section 3 of <xref target="RFC3432"/>Parameters.</t> <t><xref target="RFC3432" sectionFormat="of" section="3"/> prescribes the method for generating Periodic streams using associatedparameters.</t> <t><list style="hanging"> <t hangText="incT">theParameters.</t> <dl newline="false" spacing="normal"> <dt>incT:</dt> <dd>The nominal duration of the inter-packet interval, first bit to first bit, with value 0.0200, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1ms).</t> <t hangText="dT">thems).</dd> <dt>dT:</dt> <dd>The duration of the interval for allowed sample start times, with value 1.0, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1ms).</t> </list>NOTE: anms).</dd> </dl> <t indent="3">Note: An initiation process with a number of control exchanges resulting in unpredictable start times (within a time interval) may be sufficient to avoid synchronization of periodicstreams,streams andthereforeis a valid replacement for selecting a start time at random from a fixed interval.</t> <t>The T0parameterParameter will be reported as a measuredparameter.Parameter. Parameters incT and dT are Fixed Parameters.</t> </section> <sectiontitle="Trafficnumbered="true" toc="default"> <name>Traffic Filtering(observation) Details"> <t>NA</t>(Observation) Details</name> <t>N/A</t> </section> <sectiontitle="Sampling Distribution"> <t>NA</t>numbered="true" toc="default"> <name>Sampling Distribution</name> <t>N/A</t> </section> <sectiontitle="Run-timenumbered="true" toc="default"> <name>Runtime Parameters and DataFormat"> <t><list style="hanging"> <t hangText="Src">theFormat</name> <dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>The IP address of the host in the Src Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seeSection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="Dst">thetarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>Dst:</dt> <dd>The IP address of the host in the Dst Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seesection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="T0">atarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>T0:</dt> <dd>A time, the start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros", a start time is unspecified and Tf is to be interpreted as theDurationduration of the measurement interval. The start time is controlled through othermeans.</t> <t hangText="Tf">ameans.</dd> <dt>Tf:</dt> <dd>A time, the end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros",a endan ending time and date is ignored and Tf is interpreted as theDurationduration of the measurementinterval.</t> </list></t> <t/>interval.</dd> </dl> </section> <sectiontitle="Roles"> <t><list style="hanging"> <t hangText="Src">launchesnumbered="true" toc="default"> <name>Roles</name> <dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>Launches each packet and waits for return transmissions fromDst.</t> <t hangText="Dst">waitsthe Dst.</dd> <dt>Dst:</dt> <dd>Waits for each packet from the Src and sends a return packet toSrc.</t> </list></t>the Src (when required by the test protocol).</dd> </dl> </section> </section> <sectiontitle="Output">numbered="true" toc="default"> <name>Output</name> <t>This category specifies all details of theOutputoutput of measurements using the metric.</t> <sectiontitle="Type"> <t>Percentile -- fornumbered="true" toc="default"> <name>Type</name> <t>Percentile: For the conditional distribution of all packets with a valid value of one-way delay (undefined delays are excluded), this is a single value corresponding to the 95th percentile, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>The percentile = 95, meaning that the reported delay, "95Percentile", is the smallest value of one-way PDV for which the Empirical DistributionFunction (EDF), F(95Percentile) >=Function, EDF(95Percentile), is greater than or equal to 95% of the singleton one-way PDV values in the conditional distribution. Seesection 11.3 of<xreftarget="RFC2330"/>target="RFC2330" sectionFormat="of" section="11.3"/> for the definition of the percentile statistic using the EDF.</t> </section> <sectiontitle="Reference Definition"> <t><list style="hanging"> <t hangText="T0">thenumbered="true" toc="default"> <name>Reference Definition</name> <dl newline="false" spacing="normal"> <dt>T0:</dt> <dd>The start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="Tf">thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>Tf:</dt> <dd>The end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> </list></t> <t><list style="hanging"> <t hangText="95Percentile">Thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> </dl> <dl newline="false" spacing="normal"> <dt>95Percentile:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits= 9= 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <sectiontitle="Metric Units">numbered="true" toc="default"> <name>Metric Units</name> <t>The 95thPercentilepercentile of one-way PDV is expressed in seconds.</t> </section> <sectiontitle="Calibration"> <t>Section 3.7.3 of <xref target="RFC7679"/>numbered="true" toc="default"> <name>Calibration</name> <t><xref target="RFC7679" sectionFormat="of" section="3.7.3"/> provides a means to quantify the systematic and random errors of a time measurement.In-situ calibrationCalibration in-situ could be enabled with an internal loopback that includes as much of the measurement system as possible, performs address manipulation as needed, and provides some form of isolation (e.g., deterministic delay) to avoid send-receive interface contention. Some portion of the random and systematic error can be characterized in this way.</t> <t>For one-way delay measurements, the error calibration must include an assessment of the internal clock synchronization with its external reference (this internal clock is supplying timestamps for measurement). In practice, the time offsets <xreftarget="RFC5905"/>target="RFC5905" format="default"/> of clocks at both thesourceSource anddestinationDestination are needed to estimate the systematic error due to imperfect clock synchronization (the time offsets aresmoothed, thussmoothed; thus, the random variation is not usually represented in theresults).<list style="hanging"> <t hangText="time_offset">Theresults).</t> <dl newline="false" spacing="normal"> <dt>time_offset:</dt> <dd>The time value of the result is expressed in units of seconds, as a signed value of type decimal64 with fractiondigits = 9digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> <t>When a measurement controller requests a calibration measurement, the loopback is applied and the result is output in the same format as a normalmeasurementmeasurement, with an additional indication that it is a calibration result. In any measurement, the measurement functionSHOULD<bcp14>SHOULD</bcp14> report its current estimate of the time offset <xreftarget="RFC5905"/>target="RFC5905" format="default"/> as an indicator of the degree of synchronization.</t> <t>Both internal loopback calibration and clock synchronization can be used to estimate the available accuracy of the Output Metric Units. For example, repeated loopback delay measurements will reveal the portion of theOutputoutput result resolutionwhichthat is the result of systemnoise,noise and is thus inaccurate.</t> </section> </section> <sectiontitle="Administrative items"> <t/> <section title="Status">numbered="true" toc="default"> <name>Administrative Items</name> <section numbered="true" toc="default"> <name>Status</name> <t>Current</t> </section> <sectiontitle="Requester"> <t>This RFC number</t>numbered="true" toc="default"> <name>Requester</name> <t>RFC 8912</t> </section> <sectiontitle="Revision">numbered="true" toc="default"> <name>Revision</name> <t>1.0</t> </section> <sectiontitle="Revision Date"> <t>YYYY-MM-DD</t>numbered="true" toc="default"> <name>Revision Date</name> <t>2021-11-17</t> </section> </section> <sectiontitle="Commentsnumbered="true" toc="default"> <name>Comments andRemarks">Remarks</name> <t>Lost packets represent a challenge for delay variation metrics. Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> and the delay variation applicabilitystatement<xref target="RFC5481"/>statement <xref target="RFC5481" format="default"/> for extensive analysis and comparison of PDV and an alternate metric,IPDV.</t>IPDV (Inter-Packet Delay Variation).</t> </section> </section> <!-- Section 6 --> <sectiontitle="DNSanchor="dns_response_latency_and_loss" numbered="true" toc="default"> <name>DNS Response Latency and Loss RegistryEntries">Entries</name> <t>This section gives initialregistry entriesRegistry Entries for DNS Response Latency and Loss from a network user's perspective, for a specific named resource. The metric can be measured repeatedlyusingfor differentnames.named resources. <xreftarget="RFC2681">RFC 2681</xref>target="RFC2681" format="default"></xref> defines aRound-tripround-trip delay metric. We build on that metric by specifying several of the inputparametersParameters to precisely define two metrics for measuring DNS latency and loss.</t><t>Note to IANA: Each Registry "Name" below specifies a single registry entry, whose output format varies in accordance with the name.</t><t>All column entriesbesidebesides the ID, Name, Description, and Output Reference Method categories are thesame, thussame; thus, this sectionproposesdefines twoclosely-related registry entries.closely related Registry Entries. As a result, IANAis also asked to assignhas assigned corresponding URLs to each of the two NamedMetric.</t>Metrics.</t> <sectiontitle="Summary">numbered="true" toc="default"> <name>Summary</name> <t>This category includes multiple indexes to theregistry entries,Registry Entries: the element ID andmetric name.</t>Metric Name.</t> <sectiontitle="ID (Identifier)"> <t><insert numeric identifier, an integer></t>numbered="true" toc="default"> <name>ID (Identifier)</name> <t>IANAis asked to assign differenthas allocated the numericidentifiers to each ofIdentifiers 4 and 5 for the two NamedMetrics.</t>Metric Entries in <xref target="dns_response_latency_and_loss"/>. See <xref target="name612"/> for mapping to Names.</t> </section> <!-- 6.1.2 --> <sectiontitle="Name"> <t>RTDNS_Active_IP-UDP-Poisson_RFCXXXXsec6_Seconds_Raw</t> <t>RLDNS_Active_IP-UDP-Poisson_RFCXXXXsec6_Logical_Raw</t>anchor="name612" numbered="true" toc="default"> <name>Name</name> <dl> <dt>4:</dt><dd>RTDNS_Active_IP-UDP-Poisson_RFC8912sec6_Seconds_Raw</dd> <dt>5:</dt><dd>RLDNS_Active_IP-UDP-Poisson_RFC8912sec6_Logical_Raw</dd> </dl> </section> <!-- 6.1.3 --> <sectiontitle="URI">numbered="true" toc="default"> <name>URI</name> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTDNS_Active_IP-UDP-Poisson_RFC8912sec6_Seconds_Raw"/></t> <t>URL:https://www.iana.org/ ... <name></t><eref target="https://www.iana.org/performance-metrics/RLDNS_Active_IP-UDP-Poisson_RFC8912sec6_Logical_Raw"/></t> </section> <sectiontitle="Description">numbered="true" toc="default"> <name>Description</name> <t>This is a metric for DNS Response performance from a network user's perspective, for a specific named resource. The metric can be measured repeatedly using different resource names.</t><t>RTDNS: This<dl newline="false" spacing="normal"> <dt>RTDNS:</dt><dd>This metric assesses the response time, the interval from the query transmission to theresponse.</t> <t>RLDNS: Thisresponse.</dd> <dt>RLDNS:</dt><dd>This metric indicates that the response was deemed lost. In other words, the response time exceeded the maximum waitingtime.</t>time.</dd> </dl> </section> <sectiontitle="Change Controller">numbered="true" toc="default"> <name>Change Controller</name> <t>IETF</t> </section> <sectiontitle="Versionnumbered="true" toc="default"> <name>Version (of RegistryFormat)">Format)</name> <t>1.0</t> </section> </section> <sectiontitle="Metric Definition">numbered="true" toc="default"> <name>Metric Definition</name> <t>This category includes columns to prompt the entry of all necessary details related to the metric definition, including the RFC reference and values of input factors, calledfixed parameters.</t>"Fixed Parameters".</t> <!-- 6.2.1 --> <sectiontitle="Reference Definition"> <t>Mockapetris,numbered="true" toc="default"> <name>Reference Definition</name> <t>For Delay:</t> <t indent="3">Mockapetris, P., "Domain names - implementation and specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, November1987.1987, <https://www.rfc-editor.org/info/rfc1035> (andupdates)</t> <t><xrefupdates). <xref target="RFC1035"/></t><t>Almes,<t indent="3">Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip Delay Metric for IPPM", RFC 2681, DOI 10.17487/RFC2681, September1999.</t> <t><xref target="RFC2681"/></t> <t>Section 2.4 of1999, <https://www.rfc-editor.org/info/rfc2681>. <xreftarget="RFC2681"/>target="RFC2681"/></t> <t indent="3"><xref target="RFC2681" sectionFormat="of" section="2.4"/> provides the reference definition of the singleton (single value)Round-tripround-trip delay metric.Section 3.4 of<xreftarget="RFC2681"/>target="RFC2681" sectionFormat="of" section="3.4"/> provides the reference definition expanded to cover a multi-singleton sample. Note that terms such assingleton"singleton" andsample"sample" are defined inSection 11 of<xreftarget="RFC2330"/>.</t> <t>Fortarget="RFC2330" sectionFormat="of" section="11"/>.</t> <t indent="3">For DNS Response Latency, the entities in <xreftarget="RFC1035"/>target="RFC1035" format="default"/> must be mapped to <xreftarget="RFC2681"/>.target="RFC2681" format="default"/>. The Local Host with its User Program and Resolver take theroleRole of "Src", and the Foreign Name Server takes theroleRole of "Dst".</t><t>Note<t indent="3">Note that although the<xref target="RFC2681"/>definition of"Round-trip-Delayround-trip delay betweenSrcthe Source (Src) andDstthe Destination (Dst) atT"T as provided in <xref target="RFC2681" sectionFormat="of" section="2.4"/> is directionally ambiguous in the text, this metric tightens the definition further to recognize that the host in the"Src" roleSrc Role will send the first packet to"Dst",the host in the Dst Role and will ultimately receive the corresponding return packet from"Dst"the Dst (when neitherareis lost).</t><t>Morton,<t>For Loss:</t> <t indent="3">Morton, A.,"Round-trip"Round-Trip Packet Loss Metrics", RFC 6673, DOI 10.17487/RFC6673, August2012.</t> <t><xref2012, <https://www.rfc-editor.org/info/rfc6673>. <xref target="RFC6673"/></t><t>Both<t indent="3">For DNS Response Loss, the entities in <xref target="RFC1035"/> must be mapped to <xref target="RFC6673"/>. The Local Host with its User Program and Resolver take the Role of "Src", and the Foreign Name Server takes the Role of "Dst".</t> <t indent="3">Both response time andlossLoss metrics employ a maximum waiting time for received responses, so the count of lost packets to total packets sent is the basis for the loss determination as perSection 4.3 of<xreftarget="RFC6673"/>.</t>target="RFC6673" sectionFormat="of" section="4.3"/>.</t> </section> <sectiontitle="Fixed Parameters"> <t>Type-Pnumbered="true" toc="default"> <name>Fixed Parameters</name> <dl newline="true" spacing="normal"> <dt>Type-P as defined inSection 13 of<xreftarget="RFC2330"/>: <list style="symbols"> <t>IPv4target="RFC2330" sectionFormat="of" section="13"/>:</dt><dd><t/> <dl newline="true" spacing="normal"> <dt>IPv4 headervalues: <list style="symbols"> <t>DSCP: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to0</t> <t>TTL set0</dd> <dt>TTL:</dt><dd>Set to255</t> <t>Protocol: set255</dd> <dt>Protocol:</dt><dd>Set to 17(UDP)</t> </list></t> <t>IPv6(UDP)</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>IPv6 headervalues:<list style="symbols"> <t>DSCP: set to 0</t> <t>Hop Count: set to 255</t> <t>Next Header: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to 0</dd> <dt>Hop Count:</dt><dd>Set to 255</dd> <dt>Next Header:</dt><dd>Set to 17(UDP)</t> <t>Flow Label: set to zero</t> <t>Extension Headers: none</t> </list></t> <t>UDP(UDP)</dd> <dt>Flow Label:</dt><dd>Set to 0</dd> <dt>Extension Headers:</dt><dd> None</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>UDP headervalues: <list style="symbols"> <t>Source port: 53</t> <t>Destination port: 53</t> <t>Checksum: thevalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>Source port:</dt><dd>53</dd> <dt>Destination port:</dt><dd>53</dd> <dt>Checksum:</dt><dd>The checksummust<bcp14>MUST</bcp14> be calculated and the non-zero checksum included in theheader</t> </list></t> <t>Payload: Theheader</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>Payload:</dt> <dd><t>The payload contains a DNS message as defined in <xreftarget="RFC1035">RFC 1035</xref>target="RFC1035" format="default"></xref> with the followingvalues: <list style="symbols"> <t>Thevalues:</t> <dl newline="true" spacing="normal"> <dt>The DNS header sectioncontains: <list style="symbols"> <t>Identificationcontains:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>Identification (see theRun-time column)</t> <t>QR: setRuntime column)</dt><dd/> <dt>QR:</dt><dd>Set to 0(Query)</t> <t>OPCODE: set(Query)</dd> <dt>OPCODE:</dt><dd>Set to 0 (standardquery)</t> <t>AA: not set</t> <t>TC: not set</t> <t>RD: setquery)</dd> <dt>AA:</dt><dd>Not set</dd> <dt>TC:</dt><dd>Not set</dd> <dt>RD:</dt><dd>Set toone1 (recursiondesired)</t> <t>RA: not set</t> <t>RCODE: not set</t> <t>QDCOUNT: setdesired)</dd> <dt>RA:</dt><dd>Not set</dd> <dt>RCODE:</dt><dd>Not set</dd> <dt>QDCOUNT:</dt><dd>Set toone1 (only oneentry)</t> <t>ANCOUNT: not set</t> <t>NSCOUNT: not set</t> <t>ARCOUNT: not set</t> </list></t> <t>Theentry)</dd> <dt>ANCOUNT:</dt><dd>Not set</dd> <dt>NSCOUNT:</dt><dd>Not set</dd> <dt>ARCOUNT:</dt><dd>Not set</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>The Question sectioncontains: <list style="symbols"> <t>QNAME: thecontains:</dt> <dd><t/> <dl newline="false" spacing="normal"> <dt>QNAME:</dt><dd>The Fully Qualified Domain Name (FQDN) provided as input for thetest,test; see theRun-time column</t> <t>QTYPE: theRuntime column</dd> <dt>QTYPE:</dt><dd>The query type provided as input for thetest,test; see theRun-time column</t> <t>QCLASS: setRuntime column</dd> <dt>QCLASS:</dt><dd>Set to 1 forIN</t> </list></t> <t>TheIN</dd> </dl> </dd> </dl> <dl newline="false" spacing="normal"> <dt>The other sections do not contain any ResourceRecords.</t> </list></t> </list></t> <t>Other measurement parameters:<list style="symbols"> <t>Tmax: aRecords (RRs).</dt><dd/> </dl> </dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>Other measurement Parameters:</dt> <dd><t/> <dl newline="false" spacing="normal"> <dt>Tmax:</dt><dd>A loss threshold waiting time (and to help disambiguatequeries)<list style="symbols"> <t>5.0,queries). The value is 5.0, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1 ms), with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> </list></t> </list>Observation: replytarget="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </dd> </dl> <dl newline="false" spacing="normal"> <dt>Observation:</dt><dd>Reply packets will contain a DNSresponseResponse and may containRRs.</t>RRs.</dd> </dl> </section> </section> <sectiontitle="Methodnumbered="true" toc="default"> <name>Method ofMeasurement">Measurement</name> <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguousmethodsmethod for implementations.</t> <sectiontitle="Reference Method">numbered="true" toc="default"> <name>Reference Methods</name> <t>The methodology for this metric (equivalent to Type-P-Round-trip-Delay-Poisson-Stream) is defined asType-P-Round-trip-Delay-Poisson-Streaminsection 2.6 of<xreftarget="RFC2681">RFC 2681</xref>target="RFC2681" sectionFormat="of" section="2.6"/> (for singletons) andsection 3.6 of<xreftarget="RFC2681">RFC 2681</xref>target="RFC2681" sectionFormat="of" section="3.6"/> (for samples) using the Type-P and Timeout definedunderin the FixedParameters.</t>Parameters column.</t> <t>The reference method distinguishes between long-delayed packets and lost packets by implementing a maximum waiting time for packet arrival. Tmax is the waiting time used as the threshold to declare a response packet lost. Lost packetsSHALL<bcp14>SHALL</bcp14> be designated as having undefined delay and counted for the RLDNS metric.</t> <t>The calculations on the delay (RTT)SHALL<bcp14>SHALL</bcp14> be performed on the conditional distribution, conditioned on successful packet arrival within Tmax. Also, when all packet delays are stored, the processwhichthat calculates the RTT valueMUST<bcp14>MUST</bcp14> enforce the Tmax threshold on stored values before calculations. Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>The reference method requires some way to distinguish between different packets in a stream to establish correspondence between sending times and receiving times for eachsuccessfully-arrivingsuccessfully arriving reply.</t> <t>DNSMessagesmessages bearingQueriesqueries provide for random ID Numbers in the Identification header field, so more than one query may be launched while a previous request is outstanding when the ID Number is used. Therefore, the ID NumberMUST<bcp14>MUST</bcp14> be retained at the Src and included with each response packet to disambiguate packet reordering if it occurs.</t><t>IF<t>If a DNSresponseResponse does not arrive within Tmax, the response time RTDNS is undefined, and RLDNS = 1. The Message IDSHALL<bcp14>SHALL</bcp14> be used to disambiguate the successive queries that are otherwise identical.</t> <t>Since the ID Number field is only 16 bits in length, it places a limit on the number of simultaneous outstanding DNS queries during a stress test from a single Src address.</t> <t>Refer toSection 4.4 of<xreftarget="RFC6673"/>target="RFC6673" sectionFormat="of" section="4.4"/> for an expanded discussion of the instruction to "send a Type-P packet back to the Src as quickly as possible" inSection 2.6 of<xreftarget="RFC2681">RFC 2681</xref>.target="RFC2681" sectionFormat="of" section="2.6"/>. However, the DNSServerserver is expected to perform all required functions to prepare and send a response, so the response time will include processing time and network delay.Section 8 of<xreftarget="RFC6673"/>target="RFC6673" sectionFormat="of" section="8"/> presents additional requirementswhich SHALLthat <bcp14>SHALL</bcp14> be included in themethodMethod ofmeasurementMeasurement for this metric.</t> <t>In addition to operations described in <xreftarget="RFC2681"/>,target="RFC2681" format="default"/>, the SrcMUST<bcp14>MUST</bcp14> parse the DNS headers of the reply and prepare the query response information for subsequent reporting as a measured result, along with theRound-Trip Delay.</t>round-trip delay.</t> </section> <sectiontitle="Packetnumbered="true" toc="default"> <name>Packet StreamGeneration">Generation</name> <t>This sectiongives theprovides detailsof theregarding packettraffictraffic, which is used as the basis for measurement. In IPPMmetrics,Metrics, this is called theStream, and"stream"; this stream can easily be described by providing the list of streamparameters.</t> <t>Section 11.1.3 of <xref target="RFC2330">RFC 2681</xref>Parameters.</t> <t><xref target="RFC2330" sectionFormat="of" section="11.1.3"/> provides three methods to generate Poisson sampling intervals. The reciprocal of lambda is the average packetspacing, thusspacing; thus, theRun-timeRuntime Parameter isReciprocal_lambda = 1/lambda,Reciprocal_lambda = 1&wj;/lambda, in seconds.</t> <t>Method 3is used, where<bcp14>SHALL</bcp14> be used. Where given a start time(Run-time(Runtime Parameter), the subsequent send times are all computed prior to measurement by computing thepseudo-randompseudorandom distribution of inter-packet sendtimes,times (truncating the distribution as specified in theRun-time Parameters),Parameter Trunc), and the Src sends each packet at the computed times.</t> <t>Note that Trunc is the upper limit on inter-packet times in the Poisson distribution. A random value greater than Trunc is set equal to Trunc instead.</t> </section> <sectiontitle="Trafficnumbered="true" toc="default"> <name>Traffic Filtering(observation) Details"> <t>NA</t>(Observation) Details</name> <t>N/A</t> </section> <sectiontitle="Sampling Distribution"> <t>NA</t>numbered="true" toc="default"> <name>Sampling Distribution</name> <t>N/A</t> </section> <sectiontitle="Run-timenumbered="true" toc="default"> <name>Runtime Parameters and DataFormat"> <t>Run-timeFormat</name> <t>Runtime Parameters are input factors that must be determined, configured into the measurement system, and reported with the results for the context to be complete.</t><t><list style="hanging"> <t hangText="Src">the<dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>The IP address of the host in the Src Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seeSection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="Dst">thetarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>Dst:</dt> <dd>The IP address of the host in the Dst Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seesection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="T0">atarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>T0:</dt> <dd>A time, the start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros", a start time is unspecified and Tf is to be interpreted as theDurationduration of the measurement interval. The start time is controlled through othermeans.</t> <t hangText="Tf">ameans.</dd> <dt>Tf:</dt> <dd>A time, the end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros",a endan ending time and date is ignored and Tf is interpreted as theDurationduration of the measurementinterval.</t> <t hangText="Reciprocal_lambda">averageinterval.</dd> <dt>Reciprocal_lambda:</dt> <dd>Average packet interval for PoissonStreamsstreams, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.0001 seconds (0.1 ms), and with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> <t hangText="Trunc">Uppertarget="RFC5905" sectionFormat="of" section="6"/>.</dd> <dt>Trunc:</dt> <dd>Upper limit on Poissondistributiondistribution, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.0001 seconds (0.1 ms), and with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>target="RFC5905" sectionFormat="of" section="6"/> (values above this limit will be clipped and set to the limitvalue).</t> <t hangText="ID">Thevalue).</dd> <dt>ID:</dt> <dd>The 16-bitidentifierIdentifier assigned by the program that generates thequery, and whichquery. The ID value must vary in successive queries (a list of IDs isneeded),needed); seeSection 4.1.1 of<xreftarget="RFC1035"/>.target="RFC1035" sectionFormat="of" section="4.1.1"/>. ThisidentifierIdentifier is copied into the corresponding reply and can be used by the requester (Src) tomatch-upmatch repliestowith any outstandingqueries.</t> <t hangText="QNAME">Thequeries.</dd> <dt>QNAME:</dt> <dd>The domain name of theQuery,query, formatted as specified insection 4 of<xreftarget="RFC6991"/>.</t> <t hangText="QTYPE">The Query Type,target="RFC6991" sectionFormat="of" section="4"/>.</dd> <dt>QTYPE:</dt> <dd>The query type, which will correspond to the IP address family of the query (decimal 1 for IPv4 or 28 forIPv6,IPv6), formatted as a uint16, as persection 9.2 of<xreftarget="RFC6020"/>.</t> </list></t>target="RFC6020" sectionFormat="of" section="9.2"/>.</dd> </dl> </section> <sectiontitle="Roles"> <t><list style="hanging"> <t hangText="Src">launchesnumbered="true" toc="default"> <name>Roles</name> <dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>Launches each packet and waits for return transmissions fromDst.</t> <t hangText="Dst">waitsthe Dst.</dd> <dt>Dst:</dt> <dd>Waits for each packet from the Src and sends a return packet toSrc.</t> </list></t>the Src.</dd> </dl> </section> </section> <sectiontitle="Output">numbered="true" toc="default"> <name>Output</name> <t>This category specifies all details of theOutputoutput of measurements using the metric.</t> <sectiontitle="Type"> <t>Raw -- fornumbered="true" toc="default"> <name>Type</name> <t>Raw: For each DNSQueryquery packet sent, sets of values as defined in the next column, including the status of the response, only assigning delay values to successful query-response pairs.</t> </section> <!-- 6.4.2 --> <sectiontitle="Reference Definition">numbered="true" toc="default"> <name>Reference Definition</name> <t>For all outputs:</t><t><list style="hanging"> <t hangText="T">the<dl newline="false" spacing="normal"> <dt>T:</dt> <dd>The time the DNSQueryquery was sent during the measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="dT">Thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>dT:</dt> <dd>The time value of the round-trip delay to receive the DNSresponse,Response, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref>.target="RFC5905" sectionFormat="of" section="6"/>. This value is undefined when the response packet is not received at the Src within a waiting timeTmax seconds.</t> <t hangText="Rcode">Theof Tmax seconds.</dd> <dt>RCODE:</dt> <dd>The value of theRcodeRCODE field in the DNSresponseResponse header, expressed as a uint64 as specified insection 9.2 of<xreftarget="RFC6020"/>.target="RFC6020" sectionFormat="of" section="9.2"/>. Non-zero values convey errors in the response, and such replies must be analyzed separately from successfulrequests.</t> </list></t>requests.</dd> <dt>Logical:</dt><dd>The numeric value of the result is expressed as a Logical value, where 1 = Lost and 0 = Received, as a positive value of type uint8 (represents integer values between 0 and 255, inclusively (see <xref target="RFC6020" sectionFormat="of" section="9.2"/>). Note that for queries with outcome 1 = Lost, dT and RCODE will be set to the maximum for decimal64 and uint64, respectively.</dd> </dl> </section> <!-- 6.4.3 --> <sectiontitle="Metric Units"> <t>RTDNS: Round-trip Delay,numbered="true" toc="default"> <name>Metric Units</name> <dl newline="false" spacing="normal"> <dt>RTDNS:</dt><dd>Round-trip delay, dT, is expressed inseconds.</t> <t>RTLDNS: theseconds.</dd> <dt>RLDNS:</dt><dd>The Logical value, where 1 = Lost and 0 =Received.</t>Received.</dd> </dl> </section> <sectiontitle="Calibration"> <t>Section 3.7.3 of <xref target="RFC7679"/>numbered="true" toc="default"> <name>Calibration</name> <t><xref target="RFC7679" sectionFormat="of" section="3.7.3"/> provides a means to quantify the systematic and random errors of a time measurement.In-situ calibrationCalibration in-situ could be enabled with an internal loopback at the Source host that includes as much of the measurement system as possible, performs address and payload manipulation as needed, and provides some form of isolation (e.g., deterministic delay) to avoid send-receive interface contention. Some portion of the random and systematic error can be characterized in this way.</t> <t>When a measurement controller requests a calibration measurement, the loopback is applied and the result is output in the same format as a normalmeasurementmeasurement, with an additional indication that it is a calibration result.</t> <t>Both internal loopback calibration and clock synchronization can be used to estimate the available accuracy of the Output Metric Units. For example, repeated loopback delay measurements will reveal the portion of theOutputoutput result resolutionwhichthat is the result of systemnoise,noise and is thus inaccurate.</t> </section> </section> <sectiontitle="Administrative items"> <t/> <section title="Status">numbered="true" toc="default"> <name>Administrative Items</name> <section numbered="true" toc="default"> <name>Status</name> <t>Current</t> </section> <sectiontitle="Requester"> <t>This RFC number</t>numbered="true" toc="default"> <name>Requester</name> <t>RFC 8912</t> </section> <sectiontitle="Revision">numbered="true" toc="default"> <name>Revision</name> <t>1.0</t> </section> <sectiontitle="Revision Date"> <t>YYYY-MM-DD</t>numbered="true" toc="default"> <name>Revision Date</name> <t>2021-11-17</t> </section> </section> <sectiontitle="Commentsnumbered="true" toc="default"> <name>Comments andRemarks">Remarks</name> <t>None</t> </section> </section> <!-- Section 7 --> <sectiontitle="UDPanchor="udp-poisson-owd-owl-reg" numbered="true" toc="default"> <name>UDP PoissonOne-wayOne-Way Delay and Loss RegistryEntries">Entries</name> <t>This section specifies five initialregistry entriesRegistry Entries fortheUDP PoissonOne-way Delay,One-Way Delay and one entry for UDP PoissonOne-wayOne-Way Loss.</t><t>IANA Note: Registry "Name" below specifies multiple registry entries, whose output format varies according to the <statistic> element of the name that specifies one form of statistical summary. There is an additional metric name for the Loss metric.</t><t>All column entriesbesidebesides the ID, Name, Description, and Output Reference Method categories are thesame, thussame; thus, this sectionproposesdefines sixclosely-related registry entries.closely related Registry Entries. As a result, IANAis also asked to assignhas assigned corresponding URLs to each of the NamedMetric.</t>Metrics.</t> <sectiontitle="Summary">numbered="true" toc="default"> <!-- section 7 --> <name>Summary</name> <t>This category includes multiple indexes to theregistry entries,Registry Entries: the element ID andmetric name.</t>Metric Name.</t> <sectiontitle="ID (Identifier)">numbered="true" toc="default"> <!-- section 7.1.1 --> <name>ID (Identifier)</name> <t>IANAis asked to assign differenthas allocated the numericidentifiers to each ofIdentifiers 6-11 for the sixMetrics.</t>Named Metric Entries in <xref target="udp-poisson-owd-owl-reg"/>. See <xref target="name712"/> for mapping to Names.</t> </section> <!-- section 7.1.2 --> <sectiontitle="Name"> <t>OWDelay_Active_IP-UDP-Poisson-Payload250B_RFCXXXXsec7_Seconds_<statistic></t> <t>where <statistic> is one of:</t> <t><list style="symbols"> <t>95Percentile</t> <t>Mean</t> <t>Min</t> <t>Max</t> <t>StdDev</t> </list></t> <t>OWLoss_Active_IP-UDP-Poisson-Payload250B_RFCXXXXsec7_Percent_LossRatio</t>anchor="name712" numbered="true" toc="default"> <name>Name</name> <dl spacing="normal" newline="false" indent="5"> <dt>6:</dt><dd>OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_95Percentile</dd> <dt>7:</dt><dd>OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_Mean</dd> <dt>8:</dt><dd>OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_Min</dd> <dt>9:</dt><dd>OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_Max</dd> <dt>10:</dt><dd>OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_StdDev</dd> <dt>11:</dt><dd>OWLoss_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Percent_LossRatio</dd> </dl> </section> <!-- section 7.1.3 --> <sectiontitle="URI">numbered="true" toc="default"> <name>URI</name> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_95Percentile" /></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_Mean"/></t> <t>URL:https://www.iana.org/ ... <name></t><eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_Min"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_Max"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Seconds_StdDev"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWLoss_Active_IP-UDP-Poisson-Payload250B_RFC8912sec7_Percent_LossRatio"/></t> </section> <!-- section 7.1.4 --> <sectiontitle="Description"> <t>OWDelay: Thisnumbered="true" toc="default"> <name>Description</name> <dl newline="false" spacing="normal"> <dt>OWDelay:</dt><dd><t>This metric assesses the delay of a stream of packets exchanged between two hosts (or measurementpoints),points) and reports the <statistic>One-wayof one-way delay for all successfully exchanged packets based on their conditional delay distribution.</t> <t>where <statistic> is one of:</t><t><list style="symbols"> <t>95Percentile</t> <t>Mean</t> <t>Min</t> <t>Max</t> <t>StdDev</t> </list></t> <t>OWLoss: This<ul spacing="normal"> <li>95Percentile</li> <li>Mean</li> <li>Min</li> <li>Max</li> <li>StdDev</li> </ul> </dd> <dt>OWLoss:</dt><dd>This metric assesses the loss ratio of a stream of packets exchanged between two hosts (which are the two measurementpoints), and the Outputpoints). The output is theOne-wayone-way loss ratio for allsuccessfully receivedtransmitted packets expressed as apercentage.</t>percentage.</dd> </dl> </section> <!-- section 7.1.5 --> <section numbered="true" toc="default"> <name>Change Controller</name> <t>IETF</t> </section> <!-- section 7.1.6 --> <section numbered="true" toc="default"> <name>Version (of Registry Format)</name> <t>1.0</t> </section> </section> <!-- section 7.2 --> <sectiontitle="Metric Definition">numbered="true" toc="default"> <name>Metric Definition</name> <t>This category includes columns to prompt the entry of all necessary details related to the metric definition, including the RFC reference and values of input factors, calledfixed parameters.</t>"Fixed Parameters".</t> <!-- section 7.2.1 --> <sectiontitle="Reference Definition">numbered="true" toc="default"> <name>Reference Definition</name> <t>ForDelay:</t> <t>Almes,delay:</t> <t indent="3">Almes, G., Kalidindi, S., Zekauskas, M., and A. Morton, Ed., "A One-Way Delay Metric for IP Performance Metrics (IPPM)", STD 81, RFC 7679, DOI 10.17487/RFC7679, January 2016,<http://www.rfc-editor.org/info/rfc7679>.</t> <t><xref<https://www.rfc-editor.org/info/rfc7679>. <xref target="RFC7679"/></t><t>Morton, A.,<t indent="3">Morton, A. and E. Stephan,E.,"Spatial Composition of Metrics", RFC 6049, DOI 10.17487/RFC6049, January2011.</t> <t><xref target="RFC6049"/></t> <t>Section 3.4 of2011, <https://www.rfc-editor.org/info/rfc6049>. <xreftarget="RFC7679"/>target="RFC6049"/></t> <t indent="3"><xref target="RFC7679" sectionFormat="of" section="3.4"/> provides the reference definition of the singleton (single value)One-wayone-way delay metric.Section 4.4 of<xreftarget="RFC7679"/>target="RFC7679" sectionFormat="of" section="4.4"/> provides the reference definition expanded to cover a multi-value sample. Note that terms such assingleton"singleton" andsample"sample" are defined inSection 11 of<xreftarget="RFC2330"/>.</t> <t>Onlytarget="RFC2330" sectionFormat="of" section="11"/>.</t> <t indent="3">Only successful packet transfers with finite delay are included in the sample, as prescribed insection 4.1.2 of<xreftarget="RFC6049"/>.</t>target="RFC6049" sectionFormat="of" section="4.1.2"/>.</t> <t>For loss:</t><t>Almes,<t indent="3">Almes, G.,Kalidini,Kalidindi, S., Zekauskas, M., and A. Morton, Ed., "A One-Way Loss Metric for IP Performance Metrics (IPPM)", STD 82, RFC 7680, DOI 10.17487/RFC7680, January 2016,<http://www.rfc-editor.org/info/rfc7680>.</t> <t>Section 2.4 of<https://www.rfc-editor.org/info/rfc7680>. <xreftarget="RFC7680"/>target="RFC7680"/></t> <t indent="3"><xref target="RFC7680" sectionFormat="of" section="2.4"/> provides the reference definition of the singleton (single value) one-waylossLoss metric.Section 3.4 of<xreftarget="RFC7680"/>target="RFC7680" sectionFormat="of" section="3.4"/> provides the reference definition expanded to cover a multi-singleton sample. Note that terms such assingleton"singleton" andsample"sample" are defined inSection 11 of<xreftarget="RFC2330"/>.</t>target="RFC2330" sectionFormat="of" section="11"/>.</t> </section> <!-- section 7.2.2 --> <sectiontitle="Fixed Parameters"> <t>Type-P:<list style="symbols"> <t>IPv4numbered="true" toc="default"> <name>Fixed Parameters</name> <dl newline="true" spacing ="normal"> <dt>Type-P:</dt> <dd><t/> <dl newline="true" spacing="normal"> <dt>IPv4 headervalues: <list style="symbols"> <t>DSCP: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to0</t> <t>TTL: set0</dd> <dt>TTL:</dt><dd>Set to255</t> <t>Protocol: Set255</dd> <dt>Protocol:</dt><dd>Set to 17(UDP)</t> </list></t> <t>IPv6(UDP)</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>IPv6 headervalues:<list style="symbols"> <t>DSCP: set to 0</t> <t>Hop Count: set to 255</t> <t>Next Header: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to 0</dd> <dt>Hop Count:</dt><dd>Set to 255</dd> <dt>Next Header:</dt><dd>Set to 17(UDP)</t> <t>Flow Label: set(UDP)</dd> <dt>Flow Label:</dt><dd> Set tozero</t> <t>Extension Headers: none</t> </list></t> <t>UDP0</dd> <dt>Extension Headers:</dt><dd>None</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>UDP headervalues: <list style="symbols"> <t>Checksum: thevalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>Checksum:</dt><dd>The checksumMUST<bcp14>MUST</bcp14> be calculated and the non-zero checksum included in theheader</t> </list></t> <t>UDP Payload: TWAMP Test Packet Formats, Section 4.1.2 of <xref target="RFC5357"/> <list style="symbols"> <t>Securityheader</dd> </dl> </dd> </dl> <dl newline="false" spacing="normal"> <dt>UDP Payload:</dt><dd><t>TWAMP-Test packet formats (<xref target="RFC5357" sectionFormat="of" section="4.1.2"/>)</t> <ul empty="true"> <li>Security features in use influence the number of Paddingoctets.</t> <t>250octets</li> <li>250 octets total, including the TWAMP format type, whichMUST<bcp14>MUST</bcp14> bereported.</t> </list></t> </list></t> <t>Other measurement parameters:</t> <t><list style="hanging"> <t hangText="Tmax:">areported</li> </ul> </dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>Other measurement Parameters:</dt> <dd><t/> <dl newline="false" spacing="normal"> <dt>Tmax:</dt> <dd>A loss threshold waiting time with value 3.0, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1 ms), with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </dd> </dl> <t>See the Packet Streamgeneration categoryGeneration section for two additional Fixed Parameters.</t> </section> </section> <!-- section 7.3 --> <sectiontitle="Methodnumbered="true" toc="default"> <name>Method ofMeasurement">Measurement</name> <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguousmethodsmethod for implementations.</t> <!-- section 7.3.1 --> <sectiontitle="Reference Method">numbered="true" toc="default"> <name>Reference Methods</name> <t>The methodology for this metric (equivalent to Type-P-One-way-Delay-Poisson-Stream) is defined asType-P-One-way-Delay-Poisson-Streaminsection 3.6 of<xreftarget="RFC7679"/>target="RFC7679" sectionFormat="of" section="3.6"/> (for singletons) andsection 4.6 of<xreftarget="RFC7679"/>target="RFC7679" sectionFormat="of" section="4.6"/> (for samples) using the Type-P and Tmax definedunderin the FixedParameters.</t>Parameters column.</t> <t>The reference method distinguishes between long-delayed packets and lost packets by implementing a maximum waiting time for packet arrival. Tmax is the waiting time used as the threshold to declare a packet lost. Lost packetsSHALL<bcp14>SHALL</bcp14> be designated as having undefineddelay,delay and counted for the OWLoss metric.</t> <t>The calculations on the one-way delaySHALL<bcp14>SHALL</bcp14> be performed on the conditional distribution, conditioned on successful packet arrival within Tmax. Also, when all packet delays are stored, the processwhichthat calculates the one-way delay valueMUST<bcp14>MUST</bcp14> enforce the Tmax threshold on stored values before calculations. Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>The reference method requires some way to distinguish between different packets in a stream to establish correspondence between sending times and receiving times for eachsuccessfully-arrivingsuccessfully arriving packet.</t> <t>Since a standard measurement protocol is employed <xreftarget="RFC5357"/>, thentarget="RFC5357" format="default"/>, the measurement process will determine the sequence numbers or timestamps applied to test packets after the Fixed and RuntimeparametersParameters are passed to that process. The measurement protocol dictates the format of sequence numbers andtime-stampstimestamps conveyed in the TWAMP-Test packet payload.</t> </section> <!-- section 7.3.2 --> <sectiontitle="Packetnumbered="true" toc="default"> <name>Packet StreamGeneration">Generation</name> <t>This sectiongives theprovides detailsof theregarding packettraffictraffic, which is used as the basis for measurement. In IPPMmetrics,Metrics, this is called theStream, and"stream"; this stream can easily be described by providing the list of streamparameters.</t> <t>Section 11.1.3 of <xref target="RFC2330">RFC 2681</xref>Parameters.</t> <t><xref target="RFC2330" sectionFormat="of" section="11.1.3"/> provides three methods to generate Poisson sampling intervals. The reciprocal of lambda is the average packetspacing, thusspacing; thus, theRun-timeRuntime Parameter isReciprocal_lambda = 1/lambda,Reciprocal_lambda = 1&wj;/lambda, in seconds.</t> <t>Method 3SHALL<bcp14>SHALL</bcp14> beused, whereused. Where given a start time(Run-time(Runtime Parameter), the subsequent send times are all computed prior to measurement by computing thepseudo-randompseudorandom distribution of inter-packet sendtimes,times (truncating the distribution as specified in the Parameter Trunc), and the Src sends each packet at the computed times.</t> <t>Note that Trunc is the upper limit on inter-packet times in the Poisson distribution. A random value greater than Trunc is set equal to Trunc instead.</t><t><list style="hanging"> <t hangText="Reciprocal_lambda">average<dl newline="false" spacing="normal"> <dt>Reciprocal_lambda:</dt> <dd>Average packet interval for PoissonStreamsstreams, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.0001 seconds (0.1 ms), and with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.target="RFC5905" sectionFormat="of" section="6"/>. Reciprocal_lambda = 1second.</t> <t hangText="Trunc">Uppersecond.</dd> <dt>Trunc:</dt> <dd>Upper limit on Poissondistributiondistribution, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.0001 seconds (0.1 ms), and with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>target="RFC5905" sectionFormat="of" section="6"/> (values above this limit will be clipped and set to the limit value).Trunc = 30.0000 seconds.</t> </list></t>Trunc = 30.0000 seconds.</dd> </dl> </section> <!-- section 7.3.3 --> <sectiontitle="Trafficnumbered="true" toc="default"> <name>Traffic Filtering(observation) Details"> <t>NA</t>(Observation) Details</name> <t>N/A</t> </section> <!-- section 7.3.4 --> <sectiontitle="Sampling Distribution"> <t>NA</t>numbered="true" toc="default"> <name>Sampling Distribution</name> <t>N/A</t> </section> <!-- section 7.3.5 --> <sectiontitle="Run-timenumbered="true" toc="default"> <name>Runtime Parameters and DataFormat"> <t>Run-timeFormat</name> <t>Runtime Parameters are input factors that must be determined, configured into the measurement system, and reported with the results for the context to be complete.</t><t><list style="hanging"> <t hangText="Src">the<dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>The IP address of the host in the Src Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seeSection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="Dst">thetarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>Dst:</dt> <dd>The IP address of the host in the Dst Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seesection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="T0">atarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>T0:</dt> <dd>A time, the start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros", a start time is unspecified and Tf is to be interpreted as theDurationduration of the measurement interval. The start time is controlled through othermeans.</t> <t hangText="Tf">ameans.</dd> <dt>Tf:</dt> <dd>A time, the end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros",a endan ending time and date is ignored and Tf is interpreted as theDurationduration of the measurementinterval.</t> </list></t>interval.</dd> </dl> </section> <!-- section 7.3.6 --> <sectiontitle="Roles"> <t><list style="hanging"> <t hangText="Src">launchesnumbered="true" toc="default"> <name>Roles</name> <dl newline="false" spacing="normal"> <dt>Src:</dt> <!-- updated - 7.3.6 --> <dd>Launches each packet and waits for return transmissions from the Dst.This An example is the TWAMPSession-Sender.</t> <t hangText="Dst">waitsSession-Sender.</dd> <dt>Dst:</dt> <dd>Waits for each packet from the Src and sends a return packet to the Src.This An example is the TWAMPSession-Reflector.</t> </list></t>Session-Reflector.</dd> </dl> </section> </section> <!-- section 7.4 --> <sectiontitle="Output">numbered="true" toc="default"> <name>Output</name> <t>This category specifies all details of theOutputoutput of measurements using the metric.</t> <!-- section 7.4.1 --> <sectiontitle="Type"> <t>See subsection titles below for Types.</t>numbered="true" toc="default"> <name>Type</name> <t>Types are discussed in the subsections below.</t> </section> <!-- section 7.4.2 --> <sectiontitle="Reference Definition">numbered="true" toc="default"> <name>Reference Definition</name> <t>For all outputtypes ---<list style="hanging"> <t hangText="T0">thetypes:</t> <dl newline="false" spacing="normal"> <dt>T0:</dt> <dd>The start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="Tf">thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>Tf:</dt> <dd>The end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> </list></t>target="RFC2330" sectionFormat="of" section="6.1"/>.</dd> </dl> <t>ForLossRatio --LossRatio, the count of lost packets to total packets sent is the basis for the loss ratio calculation as perSection 4.1 of<xreftarget="RFC7680"/>.</t>target="RFC7680" sectionFormat="of" section="4.1"/>.</t> <t>For each<statistic>,<statistic> or Percent_LossRatio, one of the followingsub-sections apply:</t>subsections applies.</t> <!-- section 7.4.2.1 --> <sectiontitle="Percentile95">numbered="true" toc="default"> <name>Percentile95</name> <t>The 95th percentileSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone-way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.3"/> for details on the percentile statistic (whereRound-tripround-trip delay should be substituted for "ipdv").</t> <t>The percentile = 95, meaning that the reported delay, "95Percentile", is the smallest value of one-way delay for which the Empirical DistributionFunction (EDF), F(95Percentile) >=Function, EDF(95Percentile), is greater than or equal to 95% of the singleton one-way delay values in the conditional distribution. Seesection 11.3 of<xreftarget="RFC2330"/>target="RFC2330" sectionFormat="of" section="11.3"/> for the definition of the percentile statistic using the EDF.</t><t><list style="hanging"> <t hangText="95Percentile">The<dl newline="false" spacing="normal"> <dt>95Percentile:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- section 7.4.2.2 --> <sectiontitle="Mean">anchor="mean-sec7422" numbered="true" toc="default"> <name>Mean</name> <t>The meanSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone-way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.2.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.2.2"/> for details on calculating thisstatistic, and 4.2.3 ofstatistic; see also <xreftarget="RFC6049"/>.</t> <t><list style="hanging"> <t hangText="Mean">Thetarget="RFC6049" sectionFormat="of" section="4.2.3"/>.</t> <dl newline="false" spacing="normal"> <dt>Mean:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- section 7.4.2.3 --> <sectiontitle="Min">numbered="true" toc="default"> <name>Min</name> <t>The minimumSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone-way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.3.2"/> for details on calculating thisstatistic, and 4.3.3 ofstatistic; see also <xreftarget="RFC6049"/>.</t> <t><list style="hanging"> <t hangText="Min">Thetarget="RFC6049" sectionFormat="of" section="4.3.3"/>.</t> <dl newline="false" spacing="normal"> <dt>Min:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- section 7.4.2.4 --> <sectiontitle="Max">numbered="true" toc="default"> <name>Max</name> <t>The maximumSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone-way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.3.2"/> for a closely related method for calculating thisstatistic, and 4.3.3 ofstatistic; see also <xreftarget="RFC6049"/>.target="RFC6049" sectionFormat="of" section="4.3.3"/>. The formula is as follows:</t><t><figure> <artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Max =(FiniteDelay [j]) such(FiniteDelay[j]) ]]></artwork> <ul empty="true"> <li>such that for some index, j, where 1<=<= j<=<= NFiniteDelay[j] >= FiniteDelay[n]FiniteDelay[j] >= FiniteDelay[n] for alln]]></artwork> </figure></t> <t><list style="hanging"> <t hangText="Max">Then</li> </ul> <dl newline="false" spacing="normal"> <dt>Max:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- section 7.4.2.5 --> <sectiontitle="Std_Dev">numbered="true" toc="default"> <name>Std_Dev</name> <t>TheStd_Dev SHALLstandard deviation (Std_Dev) <bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone&nbhy;way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 6.1.4 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="6.1.4"/> for a closely related method for calculating this statistic. The formula is the classic calculation for the standard deviation of apopulation.<figure> <artwork><![CDATA[Definepopulation.</t> <t>Define Population Std_Dev_Delay asfollows: (where all packets n = 1 through N have a value for Delay[n], and MeanDelay calculated as in 7.4.2.2), and SQRT[] is the Square Root function:follows:</t> <artwork name="" type="" align="left" alt=""><![CDATA[ _ _ | N | | --- | | 1 \ 2 | Std_Dev = SQRT | ------- > (Delay[n] - MeanDelay) | | (N) / | | --- | | n = 1 | |_ _| ]]></artwork></figure></t> <t><list style="hanging"> <t hangText="Std_Dev">The<t>where all packets n = 1 through N have a value for Delay[n], MeanDelay is calculated per <xref target="mean-sec7422"/>, and SQRT[] is the Square Root function:</t> <dl newline="false" spacing="normal"> <dt>Std_Dev:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6<xref target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 7.4.2.6 --> <section> <name>Percent_LossRatio</name> <dl> <dt>Percent_LossRatio:</dt><dd>The numeric value of the result is expressed in units of lost packets to total packets times 100%, as a positive value of type decimal64 with fraction digits = 9 (see <xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.0000000001.</dd></dl> </section> </section> <!-- section 7.4.3 --> <sectiontitle="Metric Units">numbered="true" toc="default"> <name>Metric Units</name> <t>The <statistic> ofOne-way Delayone-way delay is expressed inseconds.</t>seconds, where <statistic> is one of:</t> <ul> <li>95Percentile</li> <li>Mean</li> <li>Min</li> <li>Max</li> <li>StdDev</li> </ul> <t>TheOne-way Loss Ratioone-way loss ratio is expressed as a percentage of lost packets to total packets sent.</t> </section> <!-- 7.4.4 --> <sectiontitle="Calibration"> <t>Section 3.7.3 of <xref target="RFC7679"/>numbered="true" toc="default"> <name>Calibration</name> <t><xref target="RFC7679" sectionFormat="of" section="3.7.3"/> provides a means to quantify the systematic and random errors of a time measurement.In-situ calibrationCalibration in-situ could be enabled with an internal loopback that includes as much of the measurement system as possible, performs address manipulation as needed, and provides some form of isolation (e.g., deterministic delay) to avoid send-receive interface contention. Some portion of the random and systematic error can be characterized in this way.</t> <t>For one-way delay measurements, the error calibration must include an assessment of the internal clock synchronization with its external reference (this internal clock is supplying timestamps for measurement). In practice, the time offsets <xreftarget="RFC5905"/>target="RFC5905" format="default"/> of clocks at both thesourceSource anddestinationDestination are needed to estimate the systematic error due to imperfect clock synchronization (the time offsets <xreftarget="RFC5905"/>target="RFC5905" format="default"/> aresmoothed, thussmoothed; thus, the random variation is not usually represented in theresults).<list style="hanging"> <t hangText="time_offset">Theresults).</t> <dl newline="false" spacing="normal"> <dt>time_offset:</dt> <dd>The time value of the result is expressed in units of seconds, as a signed value of type decimal64 with fractiondigits = 9digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> <t>When a measurement controller requests a calibration measurement, the loopback is applied and the result is output in the same format as a normalmeasurementmeasurement, with an additional indication that it is a calibration result. In any measurement, the measurement functionSHOULD<bcp14>SHOULD</bcp14> report its current estimate of the time offset <xreftarget="RFC5905"/>target="RFC5905" format="default"/> as an indicator of the degree of synchronization.</t> <t>Both internal loopback calibration and clock synchronization can be used to estimate the available accuracy of the Output Metric Units. For example, repeated loopback delay measurements will reveal the portion of theOutputoutput result resolutionwhichthat is the result of systemnoise,noise and is thus inaccurate.</t><t/></section> </section> <sectiontitle="Administrative items"> <t/> <section title="Status">numbered="true" toc="default"> <name>Administrative Items</name> <section numbered="true" toc="default"> <name>Status</name> <t>Current</t> </section> <sectiontitle="Requester"> <t>This RFC number</t>numbered="true" toc="default"> <name>Requester</name> <t>RFC 8912</t> </section> <sectiontitle="Revision">numbered="true" toc="default"> <name>Revision</name> <t>1.0</t> </section> <sectiontitle="Revision Date"> <t>YYYY-MM-DD</t>numbered="true" toc="default"> <name>Revision Date</name> <t>2021-11-17</t> </section> </section> <sectiontitle="Commentsnumbered="true" toc="default"> <name>Comments andRemarks">Remarks</name> <t>None</t> </section> </section> <!-- Section 8 --> <sectiontitle="UDPanchor="UDP_periodic_owd_and_loss" numbered="true" toc="default"> <name>UDP PeriodicOne-wayOne-Way Delay and Loss RegistryEntries">Entries</name> <t>This section specifies five initialregistry entriesRegistry Entries fortheUDP PeriodicOne-way Delay,One-Way Delay and one entry for UDP PeriodicOne-wayOne-Way Loss.</t><t>IANA Note: Registry "Name" below specifies multiple registry entries, whose output format varies according to the <statistic> element of the name that specifies one form of statistical summary. There is an additional metric name for the Loss metric.</t><t>All column entriesbesidebesides the ID, Name, Description, and Output Reference Method categories are thesame, thussame; thus, this sectionproposesdefines sixclosely-related registry entries.closely related Registry Entries. As a result, IANAis also asked to assignhas assigned corresponding URLs to each of the six NamedMetric.</t>Metrics.</t> <sectiontitle="Summary">numbered="true" toc="default"> <name>Summary</name> <t>This category includes multiple indexes to theregistry entries,Registry Entries: the element ID andmetric name.</t>Metric Name.</t> <sectiontitle="ID (Identifier)">numbered="true" toc="default"> <name>ID (Identifier)</name> <t>IANAis asked to assign a differenthas allocated the numericidentifiers to each ofIdentifiers 12-17 for the sixMetrics.</t>Named Metric Entries in <xref target="UDP_periodic_owd_and_loss"/>. See <xref target="name812"/> for mapping to Names.</t> </section> <sectiontitle="Name"> <t>OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFCXXXXsec8_Seconds_<statistic></t> <t>where <statistic> is one of:</t> <t><list style="symbols"> <t>95Percentile</t> <t>Mean</t> <t>Min</t> <t>Max</t> <t>StdDev</t> </list></t> <t>OWLoss_Active_IP-UDP-Periodic-Payload142B_RFCXXXXsec8_Percent_LossRatio</t>anchor="name812" numbered="true" toc="default"> <name>Name</name> <dl spacing="normal" indent="5" newline="false"> <!-- 8.1.2 --> <dt>12:</dt><dd>OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_95Percentile</dd> <dt>13:</dt><dd>OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_Mean</dd> <dt>14:</dt><dd>OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_Min</dd> <dt>15:</dt><dd>OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_Max</dd> <dt>16:</dt><dd>OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_StdDev</dd> <dt>17:</dt><dd>OWLoss_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Percent_LossRatio</dd> </dl> </section> <sectiontitle="URI">numbered="true" toc="default"> <!-- 8.1.3 --> <name>URI</name> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_95Percentile"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_Mean"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_Min"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_Max"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/OWDelay_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Seconds_StdDev"/></t> <t>URL:https://www.iana.org/ ... <name></t><eref target="https://www.iana.org/performance-metrics/OWLoss_Active_IP-UDP-Periodic20m-Payload142B_RFC8912sec8_Percent_LossRatio"/></t> </section> <!-- 8.1.4 --> <sectiontitle="Description"> <t>OWDelay: Thisnumbered="true" toc="default"> <name>Description</name> <dl newline="false" spacing="normal"> <dt>OWDelay:</dt><dd><t>This metric assesses the delay of a stream of packets exchanged between two hosts (or measurementpoints),points) and reports the <statistic>One-wayof one-way delay for all successfully exchanged packets based on their conditional delay distribution.</t> <t>where <statistic> is one of:</t><t><list style="symbols"> <t>95Percentile</t> <t>Mean</t> <t>Min</t> <t>Max</t> <t>StdDev</t> </list></t> <t>OWLoss: This<ul spacing="normal"> <li>95Percentile</li> <li>Mean</li> <li>Min</li> <li>Max</li> <li>StdDev</li> </ul> </dd> <dt>OWLoss:</dt><dd>This metric assesses the loss ratio of a stream of packets exchanged between two hosts (which are the two measurementpoints), and the Outputpoints). The output is theOne-wayone-way loss ratio for allsuccessfully receivedtransmitted packets expressed as apercentage.</t>percentage.</dd> </dl> </section> <!-- 8.1.5 --> <section numbered="true" toc="default"> <name>Change Controller</name> <t>IETF</t> </section> <sectiontitle="Metric Definition">numbered="true" toc="default"> <name>Version (of Registry Format)</name> <t>1.0</t> </section> </section> <section numbered="true" toc="default"> <name>Metric Definition</name> <t>This category includes columns to prompt the entry of all necessary details related to the metric definition, including the RFC reference and values of input factors, calledfixed parameters.</t>"Fixed Parameters".</t> <sectiontitle="Reference Definition">numbered="true" toc="default"> <name>Reference Definition</name> <t>ForDelay:</t> <t>Almes,delay:</t> <t indent="3">Almes, G., Kalidindi, S., Zekauskas, M., and A. Morton, Ed., "A One-Way Delay Metric for IP Performance Metrics (IPPM)", STD 81, RFC 7679, DOI 10.17487/RFC7679, January 2016,<http://www.rfc-editor.org/info/rfc7679>.</t> <t><xref<https://www.rfc-editor.org/info/rfc7679>. <xref target="RFC7679"/></t><t>Morton, A.,<t indent="3">Morton, A. and E. Stephan,E.,"Spatial Composition of Metrics", RFC 6049, DOI 10.17487/RFC6049, January2011.</t> <t><xref target="RFC6049"/></t> <t>Section 3.4 of2011, <https://www.rfc-editor.org/info/rfc6049>. <xreftarget="RFC7679"/>target="RFC6049"/></t> <t indent="3"><xref target="RFC7679" sectionFormat="of" section="3.4"/> provides the reference definition of the singleton (single value)One-wayone-way delay metric.Section 4.4 of<xreftarget="RFC7679"/>target="RFC7679" sectionFormat="of" section="4.4"/> provides the reference definition expanded to cover a multi-value sample. Note that terms such assingleton"singleton" andsample"sample" are defined inSection 11 of<xreftarget="RFC2330"/>.</t> <t>Onlytarget="RFC2330" sectionFormat="of" section="11"/>.</t> <t indent="3">Only successful packet transfers with finite delay are included in the sample, as prescribed insection 4.1.2 of<xreftarget="RFC6049"/>.</t>target="RFC6049" sectionFormat="of" section="4.1.2"/>.</t> <t>For loss:</t><t>Almes,<t indent="3">Almes, G.,Kalidini,Kalidindi, S., Zekauskas, M., and A. Morton, Ed., "A One-Way Loss Metric for IP Performance Metrics (IPPM)", STD 82, RFC 7680, DOI 10.17487/RFC7680, January 2016,<http://www.rfc-editor.org/info/rfc7680>.</t> <t>Section 2.4 of<https://www.rfc-editor.org/info/rfc7680>. <xreftarget="RFC7680"/>target="RFC7680"/></t> <t indent="3"><xref target="RFC7680" sectionFormat="of" section="2.4"/> provides the reference definition of the singleton (single value) one-waylossLoss metric.Section 3.4 of<xreftarget="RFC7680"/>target="RFC7680" sectionFormat="of" section="3.4"/> provides the reference definition expanded to cover a multi-singleton sample. Note that terms such assingleton"singleton" andsample"sample" are defined inSection 11 of<xreftarget="RFC2330"/>.</t>target="RFC2330" sectionFormat="of" section="11"/>.</t> </section> <!-- 8.2.2 --> <sectiontitle="Fixed Parameters"> <t>Type-P: <list style="symbols"> <t>IPv4numbered="true" toc="default"> <name>Fixed Parameters</name> <dl newline="true" spacing="normal"> <dt>Type-P:</dt> <dd><t/> <dl newline="true" spacing="normal"> <dt>IPv4 headervalues: <list style="symbols"> <t>DSCP: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to0</t> <t>TTL: set0</dd> <dt>TTL:</dt><dd>Set to255</t> <t>Protocol: Set255</dd> <dt>Protocol:</dt><dd>Set to 17(UDP)</t> </list></t> <t>IPv6(UDP)</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>IPv6 headervalues:<list style="symbols"> <t>DSCP: set to 0</t> <t>Hop Count: set to 255</t> <t>Next Header: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to 0</dd> <dt>Hop Count:</dt><dd>Set to 255</dd> <dt>Next Header:</dt><dd>Set to 17(UDP)</t> <t>Flow Label: set to zero</t> <t>Extension Headers: none</t> </list></t> <t>UDP(UDP)</dd> <dt>Flow Label:</dt><dd>Set to 0</dd> <dt>Extension Headers:</dt><dd>None</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>UDP headervalues: <list style="symbols"> <t>Checksum: thevalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>Checksum:</dt><dd>The checksumMUST<bcp14>MUST</bcp14> be calculated and the non-zero checksum included in theheader</t> </list></t> <t>UDP Payload: TWAMP Test Packet Formats, Section 4.1.2 of <xref target="RFC5357"/> <list style="symbols"> <t>Securityheader</dd> </dl> </dd> </dl> <dl newline="false" spacing="normal"> <dt>UDP Payload:</dt><dd><t>TWAMP-Test packet formats (<xref target="RFC5357" sectionFormat="of" section="4.1.2"/>)</t> <ul empty="true"> <li>Security features in use influence the number of Paddingoctets.</t> <t>142octets</li> <li>142 octets total, including the TWAMP format (and format typeMUST<bcp14>MUST</bcp14> be reported, ifused)</t> </list></t> </list></t> <t>Other measurement parameters:</t> <t><list style="hanging"> <t hangText="Tmax:">aused)</li> </ul> </dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>Other measurement Parameters:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>Tmax:</dt> <dd>A loss threshold waiting time with value 3.0, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1 ms), with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </dd> </dl> <t>See the Packet Streamgeneration categoryGeneration section fortwothree additional Fixed Parameters.</t> </section> </section> <!-- 8.3 --> <sectiontitle="Methodnumbered="true" toc="default"> <name>Method ofMeasurement">Measurement</name> <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguousmethodsmethod for implementations.</t> <sectiontitle="Reference Method">numbered="true" toc="default"> <name>Reference Methods</name> <t>The methodology for this metric (equivalent to Type-P-One-way-Delay-Poisson-Stream) is defined asType-P-One-way-Delay-Poisson-Streaminsection 3.6 of<xreftarget="RFC7679"/>target="RFC7679" sectionFormat="of" section="3.6"/> (for singletons) andsection 4.6 of<xreftarget="RFC7679"/>target="RFC7679" sectionFormat="of" section="4.6"/> (for samples) using the Type-P and Tmax definedunderin the FixedParameters.Parameters column. However, a Periodic stream is used, as defined in <xreftarget="RFC3432"/>.</t>target="RFC3432" format="default"/>.</t> <t>The reference method distinguishes between long-delayed packets and lost packets by implementing a maximum waiting time for packet arrival. Tmax is the waiting time used as the threshold to declare a packet lost. Lost packetsSHALL<bcp14>SHALL</bcp14> be designated as having undefineddelay,delay and counted for the OWLoss metric.</t> <t>The calculations on the one-way delaySHALL<bcp14>SHALL</bcp14> be performed on the conditional distribution, conditioned on successful packet arrival within Tmax. Also, when all packet delays are stored, the processwhichthat calculates the one-way delay valueMUST<bcp14>MUST</bcp14> enforce the Tmax threshold on stored values before calculations. Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>The reference method requires some way to distinguish between different packets in a stream to establish correspondence between sending times and receiving times for eachsuccessfully-arrivingsuccessfully arriving packet.</t> <t>Since a standard measurement protocol is employed <xreftarget="RFC5357"/>, thentarget="RFC5357" format="default"/>, the measurement process will determine the sequence numbers or timestamps applied to test packets after the Fixed and RuntimeparametersParameters are passed to that process. The measurement protocol dictates the format of sequence numbers andtime-stampstimestamps conveyed in the TWAMP-Test packet payload.</t> </section> <sectiontitle="Packetnumbered="true" toc="default"> <name>Packet StreamGeneration">Generation</name> <t>This sectiongives theprovides detailsof theregarding packettraffictraffic, which is used as the basis for measurement. In IPPMmetrics,Metrics, this is called theStream, and"stream"; this stream can easily be described by providing the list of streamparameters.</t> <t>Section 3 of <xref target="RFC3432"/>Parameters.</t> <t><xref target="RFC3432" sectionFormat="of" section="3"/> prescribes the method for generating Periodic streams using associatedparameters.</t> <t><list style="hanging"> <t hangText="incT">theParameters.</t> <dl newline="false" spacing="normal"> <dt>incT:</dt> <dd>The nominal duration of the inter-packet interval, first bit to first bit, with value0.02000.0200, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1 ms), with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> <t hangText="dT">thetarget="RFC5905" sectionFormat="of" section="6"/>.</dd> <dt>dT:</dt> <dd>The duration of the interval for allowed sample start times, with value 1.0000, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1 ms), with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> <t hangText="T0">thetarget="RFC5905" sectionFormat="of" section="6"/>.</dd> <dt>T0:</dt> <dd>The actual start time of the periodic stream, determined from T0 anddT.</t> </list>NOTE: andT.</dd> </dl> <t indent="3">Note: An initiation process with a number of control exchanges resulting in unpredictable start times (within a time interval) may be sufficient to avoid synchronization of periodicstreams,streams andthereforeis a valid replacement for selecting a start time at random from a fixed interval.</t> <t>These streamparametersParameters will be specified asRun-time parameters.</t>Runtime Parameters.</t> </section> <sectiontitle="Trafficnumbered="true" toc="default"> <name>Traffic Filtering(observation) Details"> <t>NA</t>(Observation) Details</name> <t>N/A</t> </section> <sectiontitle="Sampling Distribution"> <t>NA</t>numbered="true" toc="default"> <name>Sampling Distribution</name> <t>N/A</t> </section> <sectiontitle="Run-timenumbered="true" toc="default"> <name>Runtime Parameters and DataFormat"> <t>Run-timeFormat</name> <t>Runtime Parameters are input factors that must be determined, configured into the measurement system, and reported with the results for the context to be complete.</t><t><list style="hanging"> <t hangText="Src">the<dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>The IP address of the host in the Src Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seeSection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="Dst">thetarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>Dst:</dt> <dd>The IP address of the host in the Dst Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seesection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="T0">atarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>T0:</dt> <dd>A time, the start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros", a start time is unspecified and Tf is to be interpreted as theDurationduration of the measurement interval. The start time is controlled through othermeans.</t> <t hangText="Tf">ameans.</dd> <dt>Tf:</dt> <dd>A time, the end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros",a endan ending time and date is ignored and Tf is interpreted as theDurationduration of the measurementinterval.</t> </list></t> <t/>interval.</dd> </dl> </section> <sectiontitle="Roles"> <t><list style="hanging"> <t hangText="Src">launchesnumbered="true" toc="default"> <name>Roles</name> <dl newline="false" spacing="normal"> <dt>Src:</dt> <!-- 8.3.6 --> <dd>Launches each packet and waits for return transmissions from the Dst.This An example is the TWAMPSession-Sender.</t> <t hangText="Dst">waitsSession-Sender.</dd> <dt>Dst:</dt> <dd>Waits for each packet from the Src and sends a return packet to the Src.This An example is the TWAMPSession-Reflector.</t> </list></t>Session-Reflector.</dd> </dl> </section> </section> <!-- 8.4 --> <sectiontitle="Output">numbered="true" toc="default"> <name>Output</name> <t>This category specifies all details of theOutputoutput of measurements using the metric.</t> <!-- 8.4.1 --> <sectiontitle="Type"> <t>See subsection titlesnumbered="true" toc="default"> <name>Type</name> <t>Latency and Loss Types are discussed inReference Definition for Latency Types.</t>the subsections below.</t> </section> <!-- 8.4.2 --> <sectiontitle="Reference Definition">numbered="true" toc="default"> <name>Reference Definition</name> <t>For all outputtypes ---<list style="hanging"> <t hangText="T0">thetypes:</t> <dl newline="false" spacing="normal"> <dt>T0:</dt> <dd>The start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="Tf">thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>Tf:</dt> <dd>The end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> </list></t>target="RFC2330" sectionFormat="of" section="6.1"/>.</dd> </dl> <t>ForLossRatio --LossRatio, the count of lost packets to total packets sent is the basis for the loss ratio calculation as perSection 4.1 of<xreftarget="RFC7680"/>.</t>target="RFC7680" sectionFormat="of" section="4.1"/>.</t> <t>For each<statistic>,<statistic> or Percent_LossRatio, one of the followingsub-sections apply:</t>subsections applies.</t> <!-- 8.4.2.1 --> <sectiontitle="Percentile95">numbered="true" toc="default"> <name>Percentile95</name> <t>The 95th percentileSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone-way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.3"/> for details on the percentile statistic (whereRound-tripround-trip delay should be substituted for "ipdv").</t> <t>The percentile = 95, meaning that the reported delay, "95Percentile", is the smallest value of one-way delay for which the Empirical DistributionFunction (EDF), F(95Percentile) >=Function, EDF(95Percentile), is greater than or equal to 95% of the singleton one-way delay values in the conditional distribution. Seesection 11.3 of<xreftarget="RFC2330"/>target="RFC2330" sectionFormat="of" section="11.3"/> for the definition of the percentile statistic using the EDF.</t><t><list style="hanging"> <t hangText="95Percentile">The<dl newline="false" spacing="normal"> <dt>95Percentile:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 8.4.2.2 --> <sectiontitle="Mean">anchor="mean-sec8422" numbered="true" toc="default"> <name>Mean</name> <t>The meanSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone-way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.2.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.2.2"/> for details on calculating thisstatistic, and 4.2.3 ofstatistic; see also <xreftarget="RFC6049"/>.</t> <t><list style="hanging"> <t hangText="Mean">Thetarget="RFC6049" sectionFormat="of" section="4.2.3"/>.</t> <dl newline="false" spacing="normal"> <dt>Mean:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 8.4.2.3 --> <sectiontitle="Min">numbered="true" toc="default"> <name>Min</name> <t>The minimumSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone-way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.3.2"/> for details on calculating thisstatistic, and 4.3.3 ofstatistic; see also <xreftarget="RFC6049"/>.</t> <t><list style="hanging"> <t hangText="Min">Thetarget="RFC6049" sectionFormat="of" section="4.3.3"/>.</t> <dl newline="false" spacing="normal"> <dt>Min:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 8.4.2.4 --> <sectiontitle="Max">numbered="true" toc="default"> <name>Max</name> <t>The maximumSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone-way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.3.2"/> for a closely related method for calculating thisstatistic, and 4.3.3 ofstatistic; see also <xreftarget="RFC6049"/>.target="RFC6049" sectionFormat="of" section="4.3.3"/>. The formula is as follows:</t><t><figure> <artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Max =(FiniteDelay [j]) such(FiniteDelay[j]) ]]></artwork> <ul empty="true"> <li>such that for some index, j, where 1<=<= j<=<= NFiniteDelay[j] >= FiniteDelay[n]FiniteDelay[j] >= FiniteDelay[n] for alln]]></artwork> </figure></t> <t><list style="hanging"> <t hangText="Max">Then</li> </ul> <dl newline="false" spacing="normal"> <dt>Max:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 8.4.2.5 --> <sectiontitle="Std_Dev"> <t>The Std_Dev SHALLnumbered="true" toc="default"> <name>Std_Dev</name> <t>Std_Dev <bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofOne-wayone&nbhy;way delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="6.1.4"/> for a closely related method for calculating thisstatistic, and 4.3.3 of <xref target="RFC6049"/>.statistic. The formula is the classic calculation for the standard deviation of a population.</t><figure> <artwork><![CDATA[Define<t>Define Population Std_Dev_Delay asfollows: (where all packets n = 1 through N have a value for Delay[n], and MeanDelay calculated as in 7.4.2.2), and SQRT[] is the Square Root function:follows:</t> <artwork name="" type="" align="left" alt=""><![CDATA[ _ _ | N | | --- | | 1 \ 2 | Std_Dev = SQRT | ------- > (Delay[n] - MeanDelay) | | (N) / | | --- | | n = 1 | |_ _| ]]></artwork></figure> <t/> <t><list style="hanging"> <t hangText="Std_Dev">The<t>where all packets n = 1 through N have a value for Delay[n], MeanDelay is calculated per <xref target="mean-sec8422"/>, and SQRT[] is the Square Root function:</t> <dl newline="false" spacing="normal"> <dt>Std_Dev:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6<xref target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 8.4.2.6 --> <section> <name>Percent_LossRatio</name> <dl> <dt>Percent_LossRatio:</dt><dd>The numeric value of the result is expressed in units of lost packets to total packets times 100%, as a positive value of type decimal64 with fraction digits = 9 (see <xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC6020" sectionFormat="of" section="9.3"/> with a resolution of 0.0000000001.</dd></dl> </section> </section> <!-- 8.4.3 --> <sectiontitle="Metric Units">numbered="true" toc="default"> <name>Metric Units</name> <t>The <statistic> ofOne-way Delayone-way delay is expressed in seconds, where <statistic> is one of:</t><t><list style="symbols"> <t>95Percentile</t> <t>Mean</t> <t>Min</t> <t>Max</t> <t>StdDev</t> </list></t><ul spacing="normal"> <li>95Percentile</li> <li>Mean</li> <li>Min</li> <li>Max</li> <li>StdDev</li> </ul> <t>TheOne-way Loss Ratioone-way loss ratio is expressed as a percentage of lost packets to total packets sent.</t> </section> <!-- 8.4.4 --> <sectiontitle="Calibration"> <t>Section 3.7.3 of <xref target="RFC7679"/>numbered="true" toc="default"> <name>Calibration</name> <t><xref target="RFC7679" sectionFormat="of" section="3.7.3"/> provides a means to quantify the systematic and random errors of a time measurement.In-situ calibrationCalibration in-situ could be enabled with an internal loopback that includes as much of the measurement system as possible, performs address manipulation as needed, and provides some form of isolation (e.g., deterministic delay) to avoid send-receive interface contention. Some portion of the random and systematic error can be characterized in this way.</t> <t>For one-way delay measurements, the error calibration must include an assessment of the internal clock synchronization with its external reference (this internal clock is supplying timestamps for measurement). In practice, the time offsets <xreftarget="RFC5905"/>target="RFC5905" format="default"/> of clocks at both thesourceSource anddestinationDestination are needed to estimate the systematic error due to imperfect clock synchronization (the time offsets <xreftarget="RFC5905"/>target="RFC5905" format="default"/> aresmoothed, thussmoothed; thus, the random variation is not usually represented in theresults).<list style="hanging"> <t hangText="time_offset">Theresults).</t> <dl newline="false" spacing="normal"> <dt>time_offset:</dt> <dd>The time value of the result is expressed in units of seconds, as a signed value of type decimal64 with fractiondigits = 9digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> <t>When a measurement controller requests a calibration measurement, the loopback is applied and the result is output in the same format as a normalmeasurementmeasurement, with an additional indication that it is a calibration result. In any measurement, the measurement functionSHOULD<bcp14>SHOULD</bcp14> report its current estimate of the time offset <xreftarget="RFC5905"/>target="RFC5905" format="default"/> as an indicator of the degree of synchronization.</t> <t>Both internal loopback calibration and clock synchronization can be used to estimate the available accuracy of the Output Metric Units. For example, repeated loopback delay measurements will reveal the portion of theOutputoutput result resolutionwhichthat is the result of systemnoise,noise and is thus inaccurate.</t><t/></section> </section> <!-- 8.5 --> <sectiontitle="Administrative items"> <t/>numbered="true" toc="default"> <name>Administrative Items</name> <!-- 8.5.1 --> <sectiontitle="Status">numbered="true" toc="default"> <name>Status</name> <t>Current</t> </section> <!-- 8.5.2 --> <sectiontitle="Requester"> <t>This RFC number</t>numbered="true" toc="default"> <name>Requester</name> <t>RFC 8912</t> </section> <!-- 8.5.3 --> <sectiontitle="Revision">numbered="true" toc="default"> <name>Revision</name> <t>1.0</t> </section> <!-- 8.5.4 --> <sectiontitle="Revision Date"> <t>YYYY-MM-DD</t>numbered="true" toc="default"> <name>Revision Date</name> <t>2021-11-17</t> </section> </section> <!-- 8.6 --> <sectiontitle="Commentsnumbered="true" toc="default"> <name>Comments andRemarks"> <t>None.</t>Remarks</name> <t>None</t> </section> </section> <!-- Section 9 --> <sectiontitle="ICMP Round-tripanchor="icmp_roundtrip_latency_and_loss" numbered="true" toc="default"> <name>ICMP Round-Trip Latency and Loss RegistryEntries">Entries</name> <t>This section specifies three initialregistry entriesRegistry Entries fortheICMPRound-trip Latency,Round&nbhy;Trip Latency and another entry for the ICMPRound-tripRound-Trip Loss Ratio.</t><t>IANA Note: Registry "Name" below specifies multiple registry entries, whose output format varies according to the <statistic> element of the name that specifies one form of statistical summary. There is an additional metric name for the Loss metric.</t><t>All column entriesbesidebesides the ID, Name, Description, and Output Reference Method categories are thesame, thussame; thus, this sectionproposes two closely-related registry entries.defines four closely related Registry Entries. As a result, IANAis also asked to assignhas assigned corresponding URLs to each of the four NamedMetric.</t>Metrics.</t> <!-- 9.1 --> <sectiontitle="Summary">numbered="true" toc="default"> <name>Summary</name> <t>This category includes multiple indexes to theregistry entry:Registry Entries: the element ID andmetric name.</t>Metric Name.</t> <!-- 9.1.1 --> <sectiontitle="ID (Identifier)">numbered="true" toc="default"> <name>ID (Identifier)</name> <t>IANAis asked to assign differenthas allocated the numericidentifiers to each ofIdentifiers 18-21 for the four NamedMetrics.</t>Metric Entries in <xref target="icmp_roundtrip_latency_and_loss"/>. See <xref target="name912"/> for mapping to Names.</t> </section> <!-- 9.1.2 --> <sectiontitle="Name"> <t>RTDelay_Active_IP-ICMP-SendOnRcv_RFCXXXXsec9_Seconds_<statistic></t> <t>where <statistic> is one of:</t> <t><list style="symbols"> <t>Mean</t> <t>Min</t> <t>Max</t> </list></t> <t>RTLoss_Active_IP-ICMP-SendOnRcv_RFCXXXXsec9_Percent_LossRatio</t>anchor="name912" numbered="true" toc="default"> <name>Name</name> <dl spacing="normal" indent="5" newline="false"> <dt>18:</dt><dd>RTDelay_Active_IP-ICMP-SendOnRcv_RFC8912sec9_Seconds_Mean</dd> <dt>19:</dt><dd>RTDelay_Active_IP-ICMP-SendOnRcv_RFC8912sec9_Seconds_Min</dd> <dt>20:</dt><dd>RTDelay_Active_IP-ICMP-SendOnRcv_RFC8912sec9_Seconds_Max</dd> <dt>21:</dt><dd>RTLoss_Active_IP-ICMP-SendOnRcv_RFC8912sec9_Percent_LossRatio</dd> </dl> </section> <!-- 9.1.3 --> <sectiontitle="URI">numbered="true" toc="default"> <name>URI</name> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTDelay_Active_IP-ICMP-SendOnRcv_RFC8912sec9_Seconds_Mean"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTDelay_Active_IP-ICMP-SendOnRcv_RFC8912sec9_Seconds_Min"/></t> <t>URL:https://www.iana.org/ ... <name></t><eref target="https://www.iana.org/performance-metrics/RTDelay_Active_IP-ICMP-SendOnRcv_RFC8912sec9_Seconds_Max"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTLoss_Active_IP-ICMP-SendOnRcv_RFC8912sec9_Percent_LossRatio"/></t> </section> <!-- 9.1.4 --> <sectiontitle="Description"> <t>RTDelay: Thisnumbered="true" toc="default"> <name>Description</name> <dl newline="false" spacing="normal"> <dt>RTDelay:</dt><dd><t>This metric assesses the delay of a stream of ICMP packets exchanged between two hosts (which are the two measurementpoints), and the Outputpoints). The output is theRound-tripround-trip delay for all successfully exchanged packets expressed as the <statistic> of their conditional delay distribution, where <statistic> is one of:</t><t><list style="symbols"> <t>Mean</t> <t>Min</t> <t>Max</t> </list></t> <t>RTLoss: This<ul spacing="normal"> <li>Mean</li> <li>Min</li> <li>Max</li> </ul> </dd> <dt>RTLoss:</dt><dd>This metric assesses the loss ratio of a stream of ICMP packets exchanged between two hosts (which are the two measurementpoints), and the Outputpoints). The output is theRound-tripround-trip loss ratio for allsuccessfully exchangedtransmitted packets expressed as apercentage.</t>percentage.</dd> </dl> </section> <!-- 9.1.5 --> <sectiontitle="Change Controller">numbered="true" toc="default"> <name>Change Controller</name> <t>IETF</t> </section> <!-- 9.6.1 --> <sectiontitle="Versionnumbered="true" toc="default"> <name>Version (of RegistryFormat)">Format)</name> <t>1.0</t> </section> </section> <!-- 9.2 --> <sectiontitle="Metric Definition">numbered="true" toc="default"> <name>Metric Definition</name> <t>This category includes columns to prompt the entry of all necessary details related to the metric definition, including the RFC reference and values of input factors, calledfixed parameters.</t>"Fixed Parameters".</t> <!-- 9.2.1 --> <sectiontitle="Reference Definition"> <t>Almes,numbered="true" toc="default"> <name>Reference Definition</name> <t>For delay:</t> <t indent="3">Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip Delay Metric for IPPM", RFC 2681, DOI 10.17487/RFC2681, September1999.</t> <t><xref target="RFC2681"/></t> <t>Section 2.4 of1999, <https://www.rfc-editor.org/info/rfc2681>. <xreftarget="RFC2681"/>target="RFC2681"/></t> <t indent="3"><xref target="RFC2681" sectionFormat="of" section="2.4"/> provides the reference definition of the singleton (single value)Round-tripround-trip delay metric.Section 3.4 of<xreftarget="RFC2681"/>target="RFC2681" sectionFormat="of" section="3.4"/> provides the reference definition expanded to cover a multi-singleton sample. Note that terms such assingleton"singleton" andsample"sample" are defined inSection 11 of<xreftarget="RFC2330"/>.</t> <t>Notetarget="RFC2330" sectionFormat="of" section="11"/>.</t> <t indent="3">Note that although the<xref target="RFC2681"/>definition of"Round-trip-Delayround-trip delay betweenSrcthe Source (Src) andDst"the Destination (Dst) as provided in <xref target="RFC2681" sectionFormat="of" section="2.4"/> is directionally ambiguous in the text, this metric tightens the definition further to recognize that the host in the"Src" roleSrc Role will send the first packet to"Dst",the host in the Dst Role and will ultimately receive the corresponding return packet from"Dst"the Dst (when neitherareis lost).</t><t>Finally,<t indent="3">Finally, note that the variable "dT" is used in <xreftarget="RFC2681"/>target="RFC2681" format="default"/> to refer to the value ofRound-tripround-trip delay in metric definitions and methods. The variable "dT" has beenre-usedreused in other IPPM literature to refer to differentquantities,quantities and cannot be used as a global variable name.</t><t>Morton,<t>For loss:</t> <t indent="3">Morton, A.,"Round-trip"Round-Trip Packet Loss Metrics", RFC 6673, DOI 10.17487/RFC6673, August2012.</t> <t><xref2012, <https://www.rfc-editor.org/info/rfc6673>. <xref target="RFC6673"/></t> <t>BothdelayDelay andlossLoss metrics employ a maximum waiting time for received packets, so the count of lost packets to total packets sent is the basis for the loss ratio calculation as perSection 6.1 of<xreftarget="RFC6673"/>.</t>target="RFC6673" sectionFormat="of" section="6.1"/>.</t> </section> <!-- 9.2.2 --> <sectiontitle="Fixed Parameters"> <t>Type-Pnumbered="true" toc="default"> <name>Fixed Parameters</name> <dl newline="true" spacing="normal"> <dt>Type-P as defined inSection 13 of<xreftarget="RFC2330"/>: <list style="symbols"> <t>IPv4target="RFC2330" sectionFormat="of" section="13"/>:</dt><dd><t/> <dl newline="true" spacing="normal"> <dt>IPv4 headervalues: <list style="symbols"> <t>DSCP: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to0</t> <t>TTL: set0</dd> <dt>TTL:</dt><dd>Set to255</t> <t>Protocol: Set255</dd> <dt>Protocol:</dt><dd>Set to 01(ICMP)</t> </list></t> <t>IPv6(ICMP)</dd> </dl> </dd> <dt>IPv6 headervalues:<list style="symbols"> <t>DSCP: set to 0</t> <t>Hop Count: set to 255</t> <t>Next Header: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to 0</dd> <dt>Hop Count:</dt><dd>Set to 255</dd> <dt>Next Header:</dt><dd>Set to 128 decimal(ICMP)</t> <t>Flow Label: set to zero</t> <t>Extension Headers: none</t> </list></t> <t>ICMP(ICMP)</dd> <dt>Flow Label:</dt><dd>Set to 0</dd> <dt>Extension Headers:</dt><dd>None</dd> </dl> </dd> <dt>ICMP headervalues: <list style="symbols"> <t>Type: 8values:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>Type:</dt><dd>8 (EchoRequest)</t> <t>Code: 0</t> <t>Checksum: theRequest)</dd> <dt>Code:</dt><dd>0</dd> <dt>Checksum:</dt><dd>The checksumMUST<bcp14>MUST</bcp14> be calculated and the non-zero checksum included in theheader</t> <t>(Identifierheader</dd> <dt>(Identifier andSequence Numbersequence number set atRun-Time)</t> </list></t> <t>ICMP Payload <list style="symbols"> <t>totalruntime)</dt><dd/> </dl> </dd> <dt>ICMP Payload:</dt> <dd>Total of 32 bytes of randominfo,information, constant pertest.</t> </list></t> </list></t> <t>Other measurement parameters:<list style="symbols"> <t>Tmax: atest</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>Other measurement Parameters:</dt> <dd><t/> <dl newline="false" spacing="normal"> <dt>Tmax:</dt> <dd>A loss threshold waitingtime<list style="symbols"> <t>3.0,time with value 3.0, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1 ms), with lossless conversion to/from the 32-bit NTP timestamp as persection 6 of<xreftarget="RFC5905"/>.</t> </list></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </dd> </dl> </section> </section> <!-- 9.3 --> <sectiontitle="Methodnumbered="true" toc="default"> <name>Method ofMeasurement">Measurement</name> <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguousmethodsmethod for implementations.</t> <!-- 9.3.1 --> <sectiontitle="Reference Method">numbered="true" toc="default"> <name>Reference Methods</name> <t>The methodology for this metric (equivalent to Type-P-Round-trip-Delay-Poisson-Stream) is defined asType-P-Round-trip-Delay-Poisson-Streaminsection 2.6 of<xreftarget="RFC2681">RFC 2681</xref>target="RFC2681" sectionFormat="of" section="2.6"/> (for singletons) andsection 3.6 of<xreftarget="RFC2681">RFC 2681</xref>target="RFC2681" sectionFormat="of" section="3.6"/> (for samples) using the Type-P and Tmax definedunderin the FixedParameters.</t>Parameters column.</t> <t>The reference method distinguishes between long-delayed packets and lost packets by implementing a maximum waiting time for packet arrival. Tmax is the waiting time used as the threshold to declare a packet lost. Lost packetsSHALL<bcp14>SHALL</bcp14> be designated as having undefineddelay,delay and counted for the RTLoss metric.</t> <t>The calculations on the delay (RTD)SHALL<bcp14>SHALL</bcp14> be performed on the conditional distribution, conditioned on successful packet arrival within Tmax. Also, when all packet delays are stored, the processwhichthat calculates the RTD valueMUST<bcp14>MUST</bcp14> enforce the Tmax threshold on stored values before calculations. Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>The reference method requires some way to distinguish between different packets in a stream to establish correspondence between sending times and receiving times for eachsuccessfully-arrivingsuccessfully arriving packet. Sequence numbers or other send-order identificationMUST<bcp14>MUST</bcp14> be retained at the Src or included with each packet to disambiguate packet reordering if it occurs.</t> <t>The measurement process will determine the sequence numbers applied to test packets after the Fixed and RuntimeparametersParameters are passed to that process. The ICMP measurement process and protocol will dictate the format of sequence numbers and otheridentifiers.</t>Identifiers.</t> <t>Refer toSection 4.4 of<xreftarget="RFC6673"/>target="RFC6673" sectionFormat="of" section="4.4"/> for an expanded discussion of the instruction to "send a Type-P packet back to the Src as quickly as possible" inSection 2.6 of <xref target="RFC2681">RFC 2681</xref>. Section 8 of<xreftarget="RFC6673"/>target="RFC2681" sectionFormat="of" section="2.6"/>. <xref target="RFC6673" sectionFormat="of" section="8"/> presents additional requirementswhich MUSTthat <bcp14>MUST</bcp14> be included in themethodMethod ofmeasurementMeasurement for this metric.</t> </section> <!-- 9.3.2 --> <sectiontitle="Packetnumbered="true" toc="default"> <name>Packet StreamGeneration">Generation</name> <t>This sectiongives theprovides detailsof theregarding packettraffictraffic, which is used as the basis for measurement. In IPPMmetrics,Metrics, this is called theStream, and"stream"; this stream can easily be described by providing the list of streamparameters.</t>Parameters.</t> <t>The ICMP metrics use a sending discipline called "SendOnRcv" or Send On Receive. This is a modification ofSection 3 of<xreftarget="RFC3432"/>,target="RFC3432" sectionFormat="of" section="3"/>, which prescribes the method for generating Periodic streams using associatedparametersParameters as defined below for this description:</t><t><list style="hanging"> <t hangText="incT">the<dl newline="false" spacing="normal"> <dt>incT:</dt> <dd>The nominal duration of the inter-packet interval, first bit to firstbit</t> <t hangText="dT">thebit.</dd> <dt>dT:</dt> <dd>The duration of the interval for allowed sample starttimes</t> </list></t>times.</dd> </dl> <t>The incT streamparameterParameter will be specified as aRun-time parameter,Runtime Parameter, and dT is not used in SendOnRcv.</t> <t>A SendOnRcv sender behaves exactly like a Periodic stream generator while all reply packets arrive with RTD < incT, and the inter-packet interval will be constant.</t> <t>If a reply packet arrives with RTD >= incT, then the inter-packet interval for the next sending time is nominally RTD.</t> <t>If a reply packet fails to arrive within Tmax, then the inter-packet interval for the next sending time is nominally Tmax.</t> <t>If an immediatesend on replySend On Reply arrival is desired, then setincT=0.</t>incT = 0.</t> </section> <!-- 9.3.3 --> <sectiontitle="Trafficnumbered="true" toc="default"> <name>Traffic Filtering(observation) Details"> <t>NA</t>(Observation) Details</name> <t>N/A</t> </section> <!-- 9.3.4 --> <sectiontitle="Sampling Distribution"> <t>NA</t>numbered="true" toc="default"> <name>Sampling Distribution</name> <t>N/A</t> </section> <!-- 9.3.5 --> <sectiontitle="Run-timenumbered="true" toc="default"> <name>Runtime Parameters and DataFormat"> <t>Run-timeFormat</name> <t>Runtime Parameters are input factors that must be determined, configured into the measurement system, and reported with the results for the context to be complete.</t><t><list style="hanging"> <t hangText="Src">the<dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>The IP address of the host in the Src Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seeSection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="Dst">thetarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>Dst:</dt> <dd>The IP address of the host in the Dst Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seesection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="incT">thetarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>incT:</dt> <dd>The nominal duration of the inter-packet interval, first bit to first bit, expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 4 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) and with a resolution of 0.0001 seconds (0.1ms).</t> <t hangText="T0">ams).</dd> <dt>T0:</dt> <dd>A time, the start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros", a start time is unspecified and Tf is to be interpreted as theDurationduration of the measurement interval. The start time is controlled through othermeans.</t> <t hangText="Count">Themeans.</dd> <dt>Count:</dt> <dd>The total count of ICMP Echo Requests to send, formatted as a uint16, as persection 9.2 of<xreftarget="RFC6020"/>.</t> </list></t> <t>(seetarget="RFC6020" sectionFormat="of" section="9.2"/>.</dd> </dl> <t>See the Packet Stream Generation section for additionalRun-time parameters)</t>Runtime Parameters.</t> </section> <!-- 9.3.6 --> <sectiontitle="Roles"> <t><list style="hanging"> <t hangText="Src">launchesnumbered="true" toc="default"> <name>Roles</name> <dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>Launches each packet and waits for return transmissions fromDst.</t> <t hangText="Dst">waitsthe Dst.</dd> <!-- Section 9.3.6 --> <dt>Dst:</dt> <dd>Waits for each packet from the Src and sends a return packet toSrc.</t> </list></t>the Src (ICMP Echo Reply, Type 0).</dd> </dl> </section> </section> <!-- 9.4 --> <sectiontitle="Output">numbered="true" toc="default"> <name>Output</name> <t>This category specifies all details of theOutputoutput of measurements using the metric.</t> <!-- 9.4.1 --> <sectiontitle="Type"> <t>See subsection titlesnumbered="true" toc="default"> <name>Type</name> <t>Latency and Loss Types are discussed inReference Definition for Latency Types.</t> <t>LossRatio -- the count of lost packets to total packets sent is the basis fortheloss ratio calculation as per Section 6.1 of <xref target="RFC6673"/>.</t>subsections below.</t> </section> <!-- 9.4.2 --> <sectiontitle="Reference Definition">numbered="true" toc="default"> <name>Reference Definition</name> <t>For all outputtypes ---<list style="hanging"> <t hangText="T0">thetypes:</t> <dl newline="false" spacing="normal"> <dt>T0:</dt> <dd>The start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="Tf">thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>Tf:</dt> <dd>The end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="TotalCount">thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>TotalCount:</dt> <dd>The count of packets actually sent by the Src to the Dst during the measurementinterval.</t> </list></t> <t>For LossRatio -- the count of lost packets to total packets sent is the basis for the loss ratio calculation as per Section 4.1 of <xref target="RFC7680"/>.</t>interval.</dd> </dl> <t>For each<statistic>,<statistic> or Percent_LossRatio, one of the followingsub-sections apply:</t>subsections applies.</t> <!-- 9.4.2.1 --> <sectiontitle="Mean">numbered="true" toc="default"> <name>Mean</name> <t>The meanSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofRound-tripround-trip delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.2.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.2.2"/> for details on calculating thisstatistic, and 4.2.3 ofstatistic; see also <xreftarget="RFC6049"/>.</t> <t><list style="hanging"> <t hangText="Mean">Thetarget="RFC6049" sectionFormat="of" section="4.2.3"/>.</t> <dl newline="false" spacing="normal"> <dt>Mean:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 9.4.2.2 --> <sectiontitle="Min">numbered="true" toc="default"> <name>Min</name> <t>The minimumSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofRound-tripround-trip delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.3.2"/> for details on calculating thisstatistic, and 4.3.3 ofstatistic; see also <xreftarget="RFC6049"/>.</t> <t><list style="hanging"> <t hangText="Min">Thetarget="RFC6049" sectionFormat="of" section="4.3.3"/>.</t> <dl newline="false" spacing="normal"> <dt>Min:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 9.4.2.3 --> <sectiontitle="Max">numbered="true" toc="default"> <name>Max</name> <t>The maximumSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofRound-tripround-trip delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.3.2"/> for a closely related method for calculating thisstatistic, and 4.3.3 ofstatistic; see also <xreftarget="RFC6049"/>.target="RFC6049" sectionFormat="of" section="4.3.3"/>. The formula is as follows:</t><t><figure> <artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Max =(FiniteDelay [j]) such(FiniteDelay[j]) ]]></artwork> <ul empty="true"> <li>such that for some index, j, where 1<=<= j<=<= NFiniteDelay[j] >= FiniteDelay[n]FiniteDelay[j] >= FiniteDelay[n] for alln]]></artwork> </figure></t> <t><list style="hanging"> <t hangText="Max">Then</li> </ul> <dl newline="false" spacing="normal"> <dt>Max:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6<xref target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 9.4.2.4 --> <section> <name>Percent_LossRatio</name> <t>For LossRatio, the count of lost packets to total packets sent is the basis for the loss ratio calculation as per <xref target="RFC7680" sectionFormat="of" section="4.1"/>.</t> <dl> <dt>Percent_LossRatio:</dt><dd>The numeric value of the result is expressed in units of lost packets to total packets times 100%, as a positive value of type decimal64 with fraction digits = 9 (see <xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.0000000001.</dd></dl> </section> </section> <!-- 9.4.3 --> <sectiontitle="Metric Units">numbered="true" toc="default"> <name>Metric Units</name> <t>The <statistic> ofRound-trip Delayround-trip delay is expressed in seconds, where <statistic> is one of:</t><t><list style="symbols"> <t>Mean</t> <t>Min</t> <t>Max</t> </list></t><ul spacing="normal"> <li>Mean</li> <li>Min</li> <li>Max</li> </ul> <t>TheRound-trip Loss Ratioround-trip loss ratio is expressed as a percentage of lost packets to total packets sent.</t> </section> <!-- 9.4.4 --> <sectiontitle="Calibration"> <t>Section 3.7.3 of <xref target="RFC7679"/>numbered="true" toc="default"> <name>Calibration</name> <t><xref target="RFC7679" sectionFormat="of" section="3.7.3"/> provides a means to quantify the systematic and random errors of a time measurement.In-situ calibrationCalibration in-situ could be enabled with an internal loopback at the Source host that includes as much of the measurement system as possible, performs address manipulation as needed, and provides some form of isolation (e.g., deterministic delay) to avoid send-receive interface contention. Some portion of the random and systematic error can be characterized in this way.</t> <t>When a measurement controller requests a calibration measurement, the loopback is applied and the result is output in the same format as a normalmeasurementmeasurement, with an additional indication that it is a calibration result.</t> <t>Both internal loopback calibration and clock synchronization can be used to estimate the available accuracy of the Output Metric Units. For example, repeated loopback delay measurements will reveal the portion of theOutputoutput result resolutionwhichthat is the result of systemnoise,noise and is thus inaccurate.</t> </section> </section> <!-- 9.5 --> <sectiontitle="Administrative items"> <t/>numbered="true" toc="default"> <name>Administrative Items</name> <!-- 9.5.1 --> <sectiontitle="Status">numbered="true" toc="default"> <name>Status</name> <t>Current</t> </section> <!-- 9.5.2 --> <sectiontitle="Requester"> <t>This RFC number</t>numbered="true" toc="default"> <name>Requester</name> <t>RFC 8912</t> </section> <!-- 9.5.3 --> <sectiontitle="Revision">numbered="true" toc="default"> <name>Revision</name> <t>1.0</t> </section> <!-- 9.5.4 --> <sectiontitle="Revision Date"> <t>YYYY-MM-DD</t>numbered="true" toc="default"> <name>Revision Date</name> <t>2021-11-17</t> </section> </section> <sectiontitle="Commentsnumbered="true" toc="default"> <name>Comments andRemarks">Remarks</name> <t>None</t> </section> </section> <sectiontitle="TCPanchor="tcp-rt-delay-loss-reg-entries" numbered="true" toc="default"> <!-- Section 10 --> <name>TCP Round-Trip Delay and Loss RegistryEntries">Entries</name> <t>This section specifiesthreefour initialregistry entriesRegistry Entries for the Passive assessment of TCP Round-Trip Delay (RTD) and another entry for the TCPRound-tripRound-Trip Loss Count.</t><t>IANA Note: Registry "Name" below specifies multiple registry entries, whose output format varies according to the <statistic> element of the name that specifies one form of statistical summary. There are two additional metric names for Singleton RT Delay and Packet Count metrics.</t><t>All column entriesbesidebesides the ID, Name, Description, and Output Reference Method categories are thesame, thussame; thus, this sectionproposesdefines fourclosely-related registry entries.closely related Registry Entries. As a result, IANAis also asked to assignhas assigned corresponding URLs to each of the four NamedMetric.</t>Metrics.</t> <sectiontitle="Summary">numbered="true" toc="default"> <!-- 10.1 --> <name>Summary</name> <t>This category includes multiple indexes to theregistry entry:Registry Entries: the element ID andmetric name.</t>Metric Name.</t> <sectiontitle="ID (Identifier)">numbered="true" toc="default"> <!-- 10.1.1 --> <name>ID (Identifier)</name> <t>IANAis asked to assign differenthas allocated the numericidentifiers to each ofIdentifiers 22-26 for thefourfive NamedMetrics.</t>Metric Entries in <xref target="tcp-rt-delay-loss-reg-entries"/>. See <xref target="name1012"/> for mapping to Names.</t> </section> <sectiontitle="Name"> <t>RTDelay_Passive_IP-TCP_RFCXXXXsec10_Seconds_<statistic></t> <t>where <statistic> is one of:</t> <t><list style="symbols"> <t>Mean</t> <t>Min</t> <t>Max</t> </list></t> <t>RTDelay_Passive_IP-TCP-HS_RFCXXXXsec10_Seconds_Singleton</t>anchor="name1012" numbered="true" toc="default"> <!-- 10.1.2 --> <name>Name</name> <dl spacing="normal" newline="false" indent="5"> <dt>22:</dt><dd>RTDelay_Passive_IP-TCP_RFC8912sec10_Seconds_Mean</dd> <dt>23:</dt><dd>RTDelay_Passive_IP-TCP_RFC8912sec10_Seconds_Min</dd> <dt>24:</dt><dd>RTDelay_Passive_IP-TCP_RFC8912sec10_Seconds_Max</dd> <dt>25:</dt><dd>RTDelay_Passive_IP-TCP-HS_RFC8912sec10_Seconds_Singleton</dd> </dl> <t>Note that amid-pointmidpoint observer only has the opportunity to compose a single RTDelay on the TCPHand Shake.</t> <t>RTLoss_Passive_IP-TCP_RFCXXXXsec10_Packet_Count</t>handshake.</t> <dl spacing="normal" newline="false" indent="5"> <dt>26:</dt><dd>RTLoss_Passive_IP-TCP_RFC8912sec10_Packet_Count</dd> </dl> </section> <sectiontitle="URI">numbered="true" toc="default"> <!-- 10.1.3 --> <name>URI</name> <t>URL:https://www.iana.org/ ... <name></t><eref target="https://www.iana.org/performance-metrics/RTDelay_Passive_IP-TCP_RFC8912sec10_Seconds_Mean" /></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTDelay_Passive_IP-TCP_RFC8912sec10_Seconds_Min"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTDelay_Passive_IP-TCP_RFC8912sec10_Seconds_Max"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTDelay_Passive_IP-TCP-HS_RFC8912sec10_Seconds_Singleton"/></t> <t>URL: <eref target="https://www.iana.org/performance-metrics/RTLoss_Passive_IP-TCP_RFC8912sec10_Packet_Count"/></t> </section> <sectiontitle="Description"> <t>RTDelay: Thisnumbered="true" toc="default"> <!-- 10.1.4 --> <name>Description</name> <dl newline="false" spacing="normal"> <dt>RTDelay:</dt><dd><t>This metric assesses the round-trip delay of TCP packets constituting a single connection, exchanged between two hosts. We consider the measurement of round-trip delay based on a single Observation Point (OP) <xreftarget="RFC7011"/>target="RFC7011" format="default"/> somewhere in the network. TheOutputoutput is theRound-tripround-trip delay for all successfully exchanged packets expressed as the <statistic> of their conditional delay distribution, where <statistic> is one of:</t><t><list style="symbols"> <t>Mean</t> <t>Min</t> <t>Max</t> </list></t> <t>RTLoss: This<ul spacing="normal"> <li>Mean</li> <li>Min</li> <li>Max</li> </ul> </dd> <dt>RTDelay Singleton:</dt><dd><t>This metric assesses the round-trip delay of TCP packets initiating a single connection (or 3-way handshake), exchanged between two hosts. We consider the measurement of round-trip delay based on a single Observation Point (OP) <xref target="RFC7011"/> somewhere in the network. The output is the single measurement of Round-trip delay, or Singleton.</t></dd> <dt>RTLoss:</dt><dd>This metric assesses the estimated loss count for TCP packets constituting a single connection, exchanged between two hosts. We consider the measurement of round-trip delay based on a singleObservation PointOP <xreftarget="RFC7011"/>target="RFC7011" format="default"/> somewhere in the network. TheOutputoutput is the estimatedLoss Countloss count for the measurementinterval.</t>interval.</dd> </dl> </section> <sectiontitle="Change Controller">numbered="true" toc="default"> <!-- 10.1.5 --> <name>Change Controller</name> <t>IETF</t> </section> <sectiontitle="Versionnumbered="true" toc="default"> <!-- 10.1.6 --> <name>Version (of RegistryFormat)">Format)</name> <t>1.0</t> </section> </section> <sectiontitle="Metric Definition">numbered="true" toc="default"> <!-- 10.2 --> <name>Metric Definition</name> <t>This category includes columns to prompt the entry of all necessary details related to the metric definition, including the RFC reference and values of input factors, calledfixed parameters.</t>"Fixed Parameters".</t> <sectiontitle="Reference Definitions"> <t>Although there is no RFC that describes passive measurement of Round-Trip Delay, the parallel definition for Active measurement is:</t>numbered="true" toc="default" anchor="s10.2.1"> <!-- 10.2.1 --> <name>Reference Definition</name> <t>Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip Delay Metric for IPPM", RFC 2681, DOI 10.17487/RFC2681, September1999.</t> <t><xref1999, <https://www.rfc-editor.org/info/rfc2681>. <xref target="RFC2681"/></t> <t>Although there is no RFC that describes Passive Measurement of round-trip delay, the parallel definition for Active Measurement is provided in <xref target="RFC2681"/>.</t> <t>This metric definition uses the term "wire time" as defined in <xref target="RFC2330" sectionFormat="of" section="10.2"/>, and the termssingleton"singleton" andsample"sample" as defined inSection 11 of<xreftarget="RFC2330"/>. (Section 2.4 of <xref target="RFC2681"/>target="RFC2330" sectionFormat="of" section="11"/>. (<xref target="RFC2681" sectionFormat="of" section="2.4"/> provides the reference definition of the singleton (single value)Round-tripround-trip delay metric.Section 3.4 of<xreftarget="RFC2681"/>target="RFC2681" sectionFormat="of" section="3.4"/> provides the reference definition expanded to cover a multi-singleton sample.)</t> <t>With theObservation PointOP <xreftarget="RFC7011"/> (OP)target="RFC7011" format="default"/> typically located between the hosts participating in the TCP connection, theRound-trip Delayround-trip delay metric requires two individual measurements between the OP and each host, such that the Spatial Composition <xreftarget="RFC6049"/>oftarget="RFC6049" format="default"/> of the measurements yields aRound-trip Delayround-trip delay singleton (we are extending the composition of one-way subpath delays to subpath round-trip delay).</t> <t>Using the direction of TCP SYN transmission to anchor the nomenclature, host A sends theSYNSYN, and host B replies with SYN-ACK during connection establishment. The direction of SYN transfer is considered the Forward direction of transmission, from A through the OP to B(Reverse(the Reverse direction is B through the OP to A).</t> <t>TrafficfiltersFilters reduce the packetstreamstreams at the OP to a Qualified bidirectional flow of packets.</t> <t>In the definitions below, Corresponding Packets are transferred in different directions and convey a common value in a TCP header field that establishes correspondence (to the extent possible). Examples may be found in the TCP timestamp fields.</t> <t>For a real number, RTD_fwd, >> theRound-trip Delayround-trip delay in the Forward direction from the OP to host B at time T' is RTD_fwd << it isREQUIRED<bcp14>REQUIRED</bcp14> that the OP observed a Qualified Packet to host B atwire-timewire time T', that host B received that packet and sent a Corresponding Packet back to host A, and the OP observed the Corresponding Packet atwire-timewire time T' + RTD_fwd.</t> <t>For a real number, RTD_rev, >> theRound-trip Delayround-trip delay in the Reverse direction from the OP to host A at time T'' is RTD_rev << it isREQUIRED<bcp14>REQUIRED</bcp14> that the OP observed a Qualified Packet to host A atwire-timewire time T'', that host A received that packet and sent a Corresponding Packet back to host B, and that the OP observed the Corresponding Packet atwire-timewire time T'' + RTD_rev.</t> <t>Ideally, the packet sent from host B to host A in both definitions aboveSHOULD<bcp14>SHOULD</bcp14> be the same packet (or, when measuring RTD_rev first, the packet from host A to host B in both definitions should be the same).</t> <t>TheREQUIRED<bcp14>REQUIRED</bcp14> Composition Function for a singleton ofRound-trip Delayround-trip delay at time T (where T is the earliest of T' and T'' above) is:</t> <t>RTDelay = RTD_fwd + RTD_rev</t> <t>Note that when the OP is located at host A or host B, one of the terms composing RTDelay will be zero or negligible.</t><t>When<t>Using the abbreviation HS to refer to the TCP handshake: when the Qualified and Corresponding Packets are a TCP-SYN and aTCP-SYN-ACK, thenTCP&nbhy;SYN-ACK, RTD_fwd == RTD_HS_fwd.</t> <t>When the Qualified and Corresponding Packets are a TCP-SYN-ACK and a TCP-ACK,thenRTD_rev == RTD_HS_rev.</t> <t>TheREQUIRED<bcp14>REQUIRED</bcp14> Composition Function for a singleton ofRound-trip Delayround-trip delay for the connectionHand Shake:</t>handshake is:</t> <t>RTDelay_HS = RTD_HS_fwd + RTD_HS_rev</t> <t>The definition ofRound-trip Loss Countround-trip loss count uses the nomenclature developed above, based on observation of the TCP header sequence numbers and storing the sequence number gaps observed. PacketLosseslosses can be inferredfrom:<list style="symbols"> <t>Out-of-order segments: TCPfrom:</t> <dl newline="false" spacing="normal"> <dt>Out-of-order segments:</dt><dd>TCP segments are transmitted with monotonically increasing sequence numbers, but these segments may be received out of order.Section 3 of<xreftarget="RFC4737"/>target="RFC4737" sectionFormat="of" section="3"/> describes the notion of "next expected" sequencenumbersnumbers, which can be adapted to TCP segments (for the purpose of detecting reordered packets). Observation of out-of-order segments indicates loss on the path prior to theOP,OP and creates agap.</t> <t>Duplicate segments: Section 2 of <xref target="RFC5560"/>gap.</dd> <dt>Duplicate segments:</dt><dd><xref target="RFC5560" sectionFormat="of" section="2"/> defines identical packets and is suitable for evaluation of TCP packets to detect duplication. Observation ofduplicate segments *withouta segment duplicates a segment previously observed (and thus no correspondinggap*observed segment gap) indicates loss on the path following the OP(because they overlap(e.g., the segment overlaps part of thedelivered sequence numbersoctet stream already observed atOP).</t> </list></t>the OP).</dd> </dl> <t>Each observation of an out-of-order or duplicate segment infers a singleton of loss, but the composition ofRound-trip Loss Countsround-trip loss counts will be conducted over a measurement intervalwhichthat is synonymous with a single TCP connection.</t> <t>With the above observations in the Forward direction over a measurement interval, the count of out-of-order and duplicate segments is defined as RTL_fwd. Comparable observations in the Reverse direction are defined as RTL_rev.</t> <t>For a measurement interval (corresponding to a single TCPconnection),connection) T0 to Tf, theREQUIRED<bcp14>REQUIRED</bcp14> Composition Function forathe two single-direction counts of inferred loss is:</t> <t>RTLoss = RTL_fwd + RTL_rev</t> </section> <sectiontitle="Fixed Parameters"> <t/> <t>Traffic Filters: <list style="symbols"> <t>IPv4numbered="true" toc="default"> <!-- 10.2.2 --> <name>Fixed Parameters</name> <dl newline="true" spacing="normal"> <dt>Traffic Filters:</dt> <dd><t/> <dl newline="true" spacing="normal"> <dt>IPv4 headervalues: <list style="symbols"> <t>DSCP: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to0</t> <t>Protocol: Set0</dd> <dt>Protocol:</dt><dd>Set to 06(TCP)</t> </list></t> <t>IPv6(TCP)</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>IPv6 headervalues:<list style="symbols"> <t>DSCP: set to 0</t> <t>Hop Count: set to 255</t> <t>Next Header: setvalues:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>DSCP:</dt><dd>Set to 0</dd> <dt>Hop Count:</dt><dd>Set to 255</dd> <dt>Next Header:</dt><dd>Set to 6(TCP)</t> <t>Flow Label: set to zero</t> <t>Extension Headers: none</t> </list></t> <t>TCP(TCP)</dd> <dt>Flow Label:</dt><dd>Set to 0</dd> <dt>Extension Headers:</dt><dd>None</dd> </dl> </dd> </dl> <dl newline="true" spacing="normal"> <dt>TCP headervalues: <list style="symbols"> <t>Flags: ACK,values:</dt> <dd><t/> <dl newline="false" spacing="compact"> <dt>Flags:</dt><dd>ACK, SYN, FIN, set asrequired</t> <t>Timestamprequired</dd> <dt>Timestamps Option(TSopt): Set <list style="symbols"> <t><xref target="RFC7323">Section 3.2 of </xref></t> </list></t> </list></t> </list></t> <t/>(TSopt):</dt><dd>Set. See <xref target="RFC7323" sectionFormat="of" section="3.2"/></dd> </dl> </dd> </dl> </dd> </dl> </section> </section> <sectiontitle="Methodnumbered="true" toc="default"> <!-- 10.3 --> <name>Method ofMeasurement">Measurement</name> <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguousmethodsmethod for implementations.</t> <sectiontitle="Reference Methods">anchor="ref-methods-10.3.1" numbered="true" toc="default"> <!-- 10.3.1 --> <name>Reference Methods</name> <t>Thefoundationfoundational methodology for this metric is defined inSection 4 of<xreftarget="RFC7323"/>target="RFC7323" sectionFormat="of" section="4"/> using theTimestamp OptionTimestamps option with modifications that allow application at a mid-pathObservation Point (OP)OP <xreftarget="RFC7011"/>.target="RFC7011" format="default"/>. Further details and applicable heuristics were derived from <xreftarget="Strowes"/>target="Strowes" format="default"/> and <xreftarget="Trammell-14"/>.</t>target="Trammell-14" format="default"/>.</t> <t>The Traffic Filter at the OP is configured to observe a single TCP connection. When theSYN, SYN-ACK, ACKSYN/SYN-ACK/ACK handshake occurs, it offers the first opportunity to measure both RTD_fwd (on the SYN to SYN-ACK pair) and RTD_rev (on the SYN-ACK to ACK pair). Label this singleton of RTDelay as RTDelay_HS (composed using theforwardForward andreverseReverse measurement pair). RTDelay_HSSHALL<bcp14>SHALL</bcp14> be treated separately from other RTDelays on data-bearing packets and their ACKs. The RTDelay_HS valueMAY<bcp14>MAY</bcp14> be used as asanityconsistency check onother Composedthe composed values ofRTDelay.</t>RTDelay for payload-bearing packets.</t> <t>Forpayload bearingpayload-bearing packets, the OP measures the time interval between observation of a packet withSequence Number s,sequence number "s" and the corresponding ACK with the sameSequencesequence number. When the payload is transferred from host A to host B, the observed interval is RTD_fwd.</t> <t>For payload-bearing packets, each observation of an out-of-order or duplicate segment infers a loss count, but the composition of round-trip loss counts will be conducted over a measurement interval that is synonymous with a single TCP connection.</t> <t>Because many data transfers are unidirectional (say, in the Forward direction from host A to host B), it is necessary to use pure ACK packets with Timestamp (TSval) andtheirpackets with the Timestamp value echo to perform a RTD_rev measurement. The time interval between observation of the ACK from B to A, and thecorresponding packetCorresponding Packet with a TimestampechoEcho Reply (TSecr) field <xref target="RFC7323"/>, is the RTD_rev.</t> <t>Delay Measurement Filtering Heuristics:</t><t>If Data<ul spacing="normal"> <li>If data payloads were transferred in both Forward and Reverse directions, then the Round-Trip Time MeasurementRulerule inSection 4.1 of<xreftarget="RFC7323"/>target="RFC7323" sectionFormat="of" section="4.1"/> could be applied. This rule essentially excludes any measurement using a packet unless it makes progress in the transfer (advances the left edge of the send window, consistent with <xreftarget="Strowes"/>).</t> <t>Atarget="Strowes" format="default"/>).</li> <li>A different heuristic from <xreftarget="Trammell-14"/>target="Trammell-14" format="default"/> is to exclude any RTD_rev that is larger than previously observed values. This would tend to exclude Reverse measurements taken when theApplicationapplication has no data ready to send, because considerable time could be added to RTD_rev from this source oferror.</t> <t>Noteerror.</li> <li>Note that the aboveHeuristicheuristic assumes that host A is sending data. Host A expecting a download would mean that this heuristic should be applied toRTD_fwd.</t> <t>TheRTD_fwd.</li> <li>The statistic calculations to summarize the delay (RTDelay)SHALL<bcp14>SHALL</bcp14> be performed on the conditional distribution, conditioned on successful Forward and Reverse measurementswhichthat follow theHeuristics.</t>heuristics.</li> </ul> <t>Method for Inferring Loss:</t><t>The<ul spacing="normal"> <li>The OP tracks sequence numbers and stores gaps for each direction of transmission, as well as thenext-expectednext expected sequence number as discussed in <xreftarget="Trammell-14"/>target="Trammell-14" format="default"/> and <xreftarget="RFC4737"/>.target="RFC4737" format="default"/>. Loss is inferred fromOut-of-orderout-of-order segments andDuplicate segments.</t>duplicate segments.</li> </ul> <t>Loss Measurement Filtering Heuristics:</t><t><xref target="Trammell-14"/><ul spacing="normal"> <li><xref target="Trammell-14" format="default"/> adds a window of evaluation based on theRTDelay.</t> <t>Distinguish Re-orderedRTDelay.</li> <li>Distinguish reordered packets fromOOOout-of-order segments due to loss, because the sequence number gap is filled during the same RTDelay window. Segments detected asre-orderedreordered according to <xreftarget="RFC4737"/> MUSTtarget="RFC4737" format="default"/> <bcp14>MUST</bcp14> reduce theLoss Countloss count inferred fromOut-of-order segments.</t> <t>Spuriousout-of-order segments.</li> <li>Spurious (unneeded) retransmissions (observed as duplicates) can also be reduced in this way, as described in <xreftarget="Trammell-14"/>.</t>target="Trammell-14" format="default"/>.</li> </ul> <t>Sources of Error:</t><t>The<ul spacing="normal"> <li>The principal source of RTDelay error is the host processing time to return a packet that defines the termination of a time interval. The heuristics above intend to mitigate these errors by excluding measurements where host processing time is a significant part of RTD_fwd orRTD_rev.</t> <t>ARTD_rev.</li> <li>A key source of RTLoss error is observation loss, as described insection 3Section 3 of <xreftarget="Trammell-14"/>.</t>target="Trammell-14"/>.</li> </ul> </section> <sectiontitle="Packetnumbered="true" toc="default"> <!-- 10.3.2 --> <name>Packet StreamGeneration"> <t>NA</t>Generation</name> <t>N/A</t> </section> <sectiontitle="Trafficnumbered="true" toc="default"> <!-- 10.3.3 --> <name>Traffic Filtering(observation) Details">(Observation) Details</name> <t>The Fixed Parameters above give a portion of the Traffic Filter. Other aspects will be supplied asRun-timeRuntime Parameters (below).</t> </section> <sectiontitle="Sampling Distribution">numbered="true" toc="default"> <!-- 10.3.4 --> <name>Sampling Distribution</name> <t>This metric requires a complete sample of all packets that qualify according to the Traffic Filter criteria.</t> </section> <sectiontitle="Run-timenumbered="true" toc="default"> <!-- 10.3.5 --> <name>Runtime Parameters and DataFormat"> <t>Run-timeFormat</name> <t>Runtime Parameters are input factors that must be determined, configured into the measurement system, and reported with the results for the context to be complete.</t><t><list style="hanging"> <t hangText="Src">the<dl newline="false" spacing="normal"> <dt>Src:</dt> <dd>The IP address of the host in the host A Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seeSection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="Dst">thetarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>Dst:</dt> <dd>The IP address of the host in the host B Role (formatipv4-address-no-zoneipv4&nbhy;address-no-zone value forIPv4,IPv4 or ipv6-address-no-zone value forIPv6,IPv6; seesection 4 of<xreftarget="RFC6991"/>)</t> <t hangText="T0">atarget="RFC6991" sectionFormat="of" section="4"/>).</dd> <dt>T0:</dt> <dd>A time, the start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. When T0 is "all-zeros", a start time is unspecified andTdTf is to be interpreted as theDurationduration of the measurement interval. The start time is controlled through othermeans.</t> <t hangText="Td">Optionally,means.</dd> <dt>Tf:</dt> <dd>Optionally, the end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>),target="RFC6991" sectionFormat="of" section="3"/>), or the duration (see T0). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. Alternatively, the end of the measurement intervalMAY<bcp14>MAY</bcp14> be controlled by the measured connection, where the second pair of FIN and ACK packets exchanged between host A and host B effectively ends theinterval.</t> <t hangText="TTLinterval.</dd> <dt>TTL or HopLimit">SetLimit:</dt> <dd>Set at desiredvalue.</t> </list></t> <t/>value.</dd> </dl> </section> <sectiontitle="Roles"> <t><list style="hanging"> <t hangText="host A">launchesnumbered="true" toc="default"> <!-- 10.3.6 --> <name>Roles</name> <dl newline="false" spacing="normal"> <dt>host A:</dt> <dd>Launches the SYN packet to open theconnection, andconnection. The Role of "host A" is synonymous withanthe IPaddress.</t> <t hangText="host B">repliesaddress used at host A.</dd> <dt>host B:</dt> <dd>Replies with the SYN-ACK packet to open theconnection, andconnection. The Role of "host B" is synonymous withanthe IPaddress.</t> </list></t>address used at host B.</dd> </dl> </section> </section> <sectiontitle="Output">numbered="true" toc="default"> <!-- 10.4 --> <name>Output</name> <t>This category specifies all details of theOutputoutput of measurements using the metric.</t> <sectiontitle="Type"> <t>See subsection titlesnumbered="true" toc="default"> <!-- 10.4.1 --> <name>Type</name> <t>RTDelay Types are discussed inReference Definition for RTDelay Types.</t> <t>For RTLoss --the subsections below.</t> <t>For RTLoss: The count of lost packets.</t> </section> <sectiontitle="Reference Definition">numbered="true" toc="default"> <!-- 10.4.2 --> <name>Reference Definition</name> <t>For all outputtypes ---<list style="hanging"> <t hangText="T0">thetypes:</t> <dl newline="false" spacing="normal"> <dt>T0:</dt> <dd>The start of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.</t> <t hangText="Tf">thetarget="RFC2330" sectionFormat="of" section="6.1"/>.</dd> <dt>Tf:</dt> <dd>The end of a measurementinterval,interval (format"date-and-time""date&nbhy;time" as specified inSection 5.6 of<xreftarget="RFC3339"/>,target="RFC3339" sectionFormat="of" section="5.6"/>; see alsoSection 3 of"date&nbhy;and&nbhy;time" in <xreftarget="RFC6991"/>).target="RFC6991" sectionFormat="of" section="3"/>). The UTC Time Zone is required bySection 6.1 of<xreftarget="RFC2330"/>.target="RFC2330" sectionFormat="of" section="6.1"/>. The end of the measurement intervalMAY<bcp14>MAY</bcp14> be controlled by the measured connection, where the second pair of FIN and ACK packets exchanged between host A and host B effectively ends theinterval.</t> <t hangText="...">...</t> </list></t> <t>For RTDelay_HS -- the Round tripinterval.</dd> <dt>RTDelay_Passive_IP-TCP-HS:</dt> <dd>The round-trip delay of theHandshake.</t> <t>For RTLoss -- thehandshake is a Singleton.</dd> <dt>RTLoss:</dt><dd>The count of lostpackets.</t>packets.</dd> </dl> <t>For each <statistic>, Singleton, or Loss Count, one of the followingsub-sections apply:</t>subsections applies.</t> <sectiontitle="Mean">numbered="true" toc="default"> <!-- 10.4.2.1 --> <name>Mean</name> <t>The meanSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofRound-tripround-trip delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.2.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.2.2"/> for details on calculating thisstatistic, and 4.2.3 ofstatistic; see also <xreftarget="RFC6049"/>.</t> <t><list style="hanging"> <t hangText="Mean">Thetarget="RFC6049" sectionFormat="of" section="4.2.3"/>.</t> <dl newline="false" spacing="normal"> <dt>Mean:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <sectiontitle="Min">numbered="true" toc="default"> <!-- 10.4.2.2 --> <name>Min</name> <t>The minimumSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofRound-tripround-trip delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.3.2"/> for details on calculating thisstatistic, and 4.3.3 ofstatistic; see also <xreftarget="RFC6049"/>.</t> <t><list style="hanging"> <t hangText="Min">Thetarget="RFC6049" sectionFormat="of" section="4.3.3"/>.</t> <dl newline="false" spacing="normal"> <dt>Min:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3 of<xreftarget="RFC6020"/>)target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of0.000000001 seconds0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6 of<xreftarget="RFC5905">RFC</xref></t> </list></t>target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <sectiontitle="Max">numbered="true" toc="default"> <!-- 10.4.2.3 --> <name>Max</name> <t>The maximumSHALL<bcp14>SHALL</bcp14> be calculated using the conditional distribution of all packets with a finite value ofRound-tripround-trip delay (undefined delays areexcluded),excluded) -- a singlevaluevalue, as follows:</t> <t>Seesection 4.1 of<xreftarget="RFC3393"/>target="RFC3393" sectionFormat="of" section="4.1"/> for details on the conditional distribution to exclude undefined values of delay, andSection 5 ofsee <xreftarget="RFC6703"/>target="RFC6703" sectionFormat="of" section="5"/> for background on this analysis choice.</t> <t>Seesection 4.3.2 of<xreftarget="RFC6049"/>target="RFC6049" sectionFormat="of" section="4.3.2"/> for a closely related method for calculating thisstatistic, and 4.3.3 ofstatistic; see also <xreftarget="RFC6049"/>.target="RFC6049" sectionFormat="of" section="4.3.3"/>. The formula is as follows:</t><t><figure> <artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Max =(FiniteDelay [j]) such(FiniteDelay[j]) ]]></artwork> <ul empty="true"> <li>such that for some index, j, where 1<=<= j<=<= NFiniteDelay[j] >= FiniteDelay[n]FiniteDelay[j] >= FiniteDelay[n] for alln]]></artwork> </figure></t> <t><list style="hanging"> <t hangText="Max">Then</li> </ul> <dl newline="false" spacing="normal"> <dt>Max:</dt> <dd>The time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (seesection 9.3<xref target="RFC6020" sectionFormat="of" section="9.3"/>) with a resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as per <xreftarget="RFC6020"/>)target="RFC5905" sectionFormat="of" section="6"/>.</dd> </dl> </section> <!-- 10.4.2.4 --> <section numbered="true" toc="default"> <name>Singleton</name> <t>The singleton SHALL be calculated using the successful RTD_fwd (on the SYN to SYN-ACK pair) and RTD_rev (on the SYN-ACK to ACK pair), see <xref target="ref-methods-10.3.1"/>.</t> <!-- <t>For RTDelay_Passive_IP-TCP-HS: The round-trip delay of the handshake.</t> removed because not in ACM's Registry Entry as of 6/27 --> <t>The singleton time value of the result is expressed in units of seconds, as a positive value of type decimal64 with fraction digits = 9 (see <xref target="RFC6020" sectionFormat="of" section="9.3"/>) with resolution of 0.000000001 seconds (1.0 ns), and with lossless conversion to/from the 64-bit NTP timestamp as persection 6<xref target="RFC5905" sectionFormat="of" section="6"/>.</t> </section> <!-- 10.4.2.5 --> <section numbered="true" toc="default"> <name>Loss Counts</name> <t>RTLoss_Passive_IP-TCP_RFC8912sec10_Packet_Count: The count of lost packets.</t> <t>Observation of an out-of-order segment or duplicate segment infers a loss count, after application of the Definitions of <xref target="s10.2.1"/> and the Loss Measurement Filtering Heuristics of <xreftarget="RFC5905">RFC</xref></t> </list></t>target="ref-methods-10.3.1"/>. The composition of round-trip loss counts will be conducted over a measurement interval that is synonymous with a single TCP connection.</t> <t>For a measurement interval (corresponding to a single TCP connection) T0 to Tf, the REQUIRED Composition Function for the two single- direction counts of inferred loss is:</t> <t>RTLoss = RTL_fwd + RTL_rev</t> <dl><dt>Packet count:</dt><dd>The numeric value of the result is expressed in units of lost packets, as a positive value of type uint64 (represents integer values between 0 and 18446744073709551615, inclusively (see Section 9.2 of [RFC6020]).</dd> </dl> </section> </section> <sectiontitle="Metric Units">numbered="true" toc="default"> <!-- 10.4.3 --> <name>Metric Units</name> <t>The <statistic> ofRound-trip Delayround-trip delay is expressed in seconds, where <statistic> is one of:</t><t><list style="symbols"> <t>Mean</t> <t>Min</t> <t>Max</t> </list></t><ul spacing="normal"> <li>Mean</li> <li>Min</li> <li>Max</li> </ul> <t>TheRound-trip Delayround-trip delay of theHand ShakeTCP handshake singleton is expressed in seconds.</t> <t>TheRound-trip Loss Countround-trip loss count is expressed as a number of packets.</t> <!-- <t>RTLoss_Passive_IP-TCP_RFC8912sec10_Packet_Count</t> <dl> <dt>Packet count:</dt><dd>The numeric value of the result is expressed in units of lost packets, as a positive value of type uint64 (represents integer values between 0 and 18446744073709551615, inclusively (see <xref target="RFC6020" sectionFormat="of" section="9.2"/>).</dd></dl> --> </section> <sectiontitle="Calibration">numbered="true" toc="default"> <!-- 10.4.4 --> <name>Calibration</name> <t>PassivemeasurementsMeasurements at an OP could be calibrated against anactive measurementActive Measurement (with loss emulation) at host A or host B, where theactive measurementActive Measurement represents theground-truth.</t>ground truth.</t> </section> </section> <sectiontitle="Administrative items"> <t/>numbered="true" toc="default"> <!-- 10.5 --> <name>Administrative Items</name> <sectiontitle="Status">numbered="true" toc="default"> <!-- 10.5.1 --> <name>Status</name> <t>Current</t> </section> <sectiontitle="Requester"> <t>This RFC number</t>numbered="true" toc="default"> <!-- 10.5.2 --> <name>Requester</name> <t>RFC 8912</t> </section> <sectiontitle="Revision">numbered="true" toc="default"> <!-- 10.5.3 --> <name>Revision</name> <t>1.0</t> </section> <sectiontitle="Revision Date"> <t>YYYY-MM-DD</t>numbered="true" toc="default"> <!-- 10.5.4 --> <name>Revision Date</name> <t>2021-11-17</t> </section> </section> <sectiontitle="Commentsnumbered="true" toc="default"> <!-- 10.6 --> <name>Comments andRemarks"> <t>None.</t>Remarks</name> <t>None</t> </section> </section> <sectiontitle="Security Considerations">numbered="true" toc="default"> <name>Security Considerations</name> <t>Theseregistry entriesRegistry Entries represent no known implications for InternetSecurity. Eachsecurity. With the exception of <xref target="RFC1035"/>, each RFC referenced above contains a Security Considerations section. Further, theLMAP FrameworkLarge-scale Measurement of Broadband Performance (LMAP) framework <xreftarget="RFC7594"/>target="RFC7594" format="default"/> provides both security and privacy considerations for measurements.</t> <t>There are potential privacy considerations for observed traffic, particularly forpassive metricsPassive Metrics as discussed insection 10.<xref target="tcp-rt-delay-loss-reg-entries"/>. An attacker that knows that its TCP connection is being measured can modify its behavior to skew the measurement results.</t> </section> <section anchor="IANA"title="IANA Considerations"> <!-- <t>Metrics previously defined in IETF were registered in the IANA IPPM METRICS REGISTRY, however this process was discontinued when the registry structure was found to be inadequate, and the registry was declared Obsolete <xref target="RFC6248"/>.</t> <t>The form of metric registration will finalized in this and other memos, and IANA Action will be requested when the initial contents of the registry are prepared.</t>-->numbered="true" toc="default"> <name>IANA Considerations</name> <t>IANAis requested to populate Thehas populated the Performance Metrics Registry defined in <xreftarget="I-D.ietf-ippm-metric-registry"/>target="RFC8911" format="default"/> with the values defined insections 4Sections <xref target="udp-rt-latency-loss-reg-entries" format="counter"/> through10.</t><xref target="tcp-rt-delay-loss-reg-entries" format="counter"/>.</t> <t>See the IANA Considerations section of <xreftarget="I-D.ietf-ippm-metric-registry"/>target="RFC8911" format="default"/> for additionalrequests andconsiderations.</t> </section> </middle> <back> <references> <name>References</name> <references> <name>Normative References</name> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.1035.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2330.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2681.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3339.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3393.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3432.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5560.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5905.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4737.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5357.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5481.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6020.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6049.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6673.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6991.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7011.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7323.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7679.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7680.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/> <!-- draft-ietf-ippm-metric-registry (RFC 8911) --> <reference anchor="RFC8911" target="https://www.rfc-editor.org/info/rfc8911"> <front> <title>Registry for Performance Metrics</title> <author fullname="Marcelo Bagnulo" initials="M." surname="Bagnulo"> <organization/> </author> <author fullname="Benoit Claise" initials="B." surname="Claise"> <organization/> </author> <author fullname="Phil Eardley" initials="P." surname="Eardley"> <organization/> </author> <author fullname="Al Morton" initials="A." surname="Morton"> <organization/> </author> <author fullname="Aamer Akhter" initials="A." surname="Akhter"> <organization/> </author> <date month="November" year="2021"/> </front> <seriesInfo name="RFC" value="8911"/> <seriesInfo name="DOI" value="10.17487/RFC8911"/> </reference> <reference anchor="Strowes" target="https://dl.acm.org/doi/10.1145/2507771.2507781"> <front> <title>Passively Measuring TCP Round-Trip Times</title> <author fullname="Stephen Strowes" initials="S." surname="Strowes"> <organization></organization> </author> <date month="October" year="2013"/> </front> <refcontent>Communications of the ACM, Vol. 56 No. 10, Pages 57-64</refcontent> <seriesInfo name="DOI" value="10.1145/2507771.2507781"/> </reference> <reference anchor="Trammell-14" target="https://link.springer.com/chapter/10.1007/978-3-642-54999-1_2"> <front> <title>Inline Data Integrity Signals for Passive Measurement</title> <author fullname="Brian Trammell" initials="B." surname="Trammell"> <organization></organization> </author> <author fullname="David Gugelmann" initials="D." surname="Gugelmann"> <organization></organization> </author> <author fullname="Nevil Brownlee" initials="N." surname="Brownlee"> <organization></organization> </author> <date month="March" year="2014"/> </front> <refcontent>In: Dainotti A., Mahanti A., Uhlig S. (eds) Traffic Monitoring and Analysis. TMA 2014. Lecture Notes in Computer Science, vol 8406. Springer, Berlin, Heidelberg</refcontent> <seriesInfo name="DOI" value="10.1007/978-3-642-54999-1_2"/> </reference> </references> <references> <name>Informative References</name> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.1242.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6390.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6703.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7594.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8126.xml"/> </references> </references> <sectiontitle="Acknowledgements">numbered="false" toc="default"> <name>Acknowledgments</name> <t>The authors thankBrian Trammell<contact fullname="Brian Trammell"/> for suggesting the term"Run-time"Runtime Parameters", which led to the distinction betweenrun-timeRuntime andfixed parametersFixed Parameters implemented in this memo, for identifying theIPFIXIP Flow Information Export (IPFIX) metric with Flow Key as an example, for suggesting the Passive TCP RTDmetricMetric and supporting references, and for many other productive suggestions. Thanks toPeter Koch,<contact fullname="Peter Koch"/>, who provided several useful suggestions for disambiguating successive DNSQueriesqueries in the DNS Response time metric.</t> <t>The authors also acknowledge the constructive reviews and helpful suggestions fromBarbara Stark, Juergen Schoenwaelder, Tim Carey, Yaakov Stein,<contact fullname="Barbara Stark"/>, <contact fullname="Juergen Schoenwaelder"/>, <contact fullname="Tim Carey"/>, <contact fullname="Yaakov Stein"/>, and participants in the LMAPworking group.Working Group. Thanks toMichelle Cotton<contact fullname="Michelle Cotton"/> for her early IANA reviews, and toAmanda Barber<contact fullname="Amanda Baber"/> for answering questions related to the presentation of theregistryRegistry and accessibility of the complete template via URL.</t> </section></middle> <back> <references title="Normative References"> <?rfc include="reference.RFC.1035"?> <?rfc include="reference.RFC.2119"?> <?rfc include="reference.RFC.2330"?> <?rfc ?> <?rfc ?> <?rfc ?> <?rfc ?> <?rfc include='reference.RFC.2681'?> <?rfc include='reference.RFC.3393'?> <?rfc include='reference.RFC.3339'?> <?rfc include='reference.RFC.3432'?> <?rfc include='reference.RFC.5560'?> <?rfc include='reference.RFC.5905'?> <?rfc include='reference.RFC.4737'?> <?rfc include='reference.RFC.5357'?> <?rfc include='reference.RFC.5481'?> <?rfc include='reference.RFC.6020'?> <?rfc include='reference.RFC.6049'?> <?rfc include='reference.RFC.6673'?> <?rfc include='reference.RFC.6991'?> <?rfc include='reference.RFC.7011'?> <?rfc include='reference.RFC.7323'?> <?rfc include='reference.RFC.7679'?> <?rfc include='reference.RFC.7680'?> <?rfc include='reference.RFC.8174'?> <reference anchor="I-D.ietf-ippm-metric-registry"> <front> <title>Registry for Performance Metrics</title> <author fullname="Marcelo Bagnulo" initials="M." surname="Bagnulo"> <organization/> </author> <author fullname="Benoit Claise" initials="B." surname="Claise"> <organization/> </author> <author fullname="Phil Eardley" initials="P." surname="Eardley"> <organization/> </author> <author fullname="Al Morton" initials="A." surname="Morton"> <organization/> </author> <date year="2019"/> </front> <seriesInfo name="Internet Draft (work in progress)" value="draft-ietf-ippm-metric-registry"/> <format type="TXT"/> </reference> <reference anchor="Strowes"> <front> <title>Passively Measuring TCP Round Trip Times, Communications of the ACM, Vol. 56 No. 10, Pages 57-64</title> <author fullname="S.Strowes" initials="S." surname="Strowes"> <organization>Communications of the ACM, Vol. 56 No. 10, Pages 57-64.</organization> </author> <date month="September" year="2013"/> </front> </reference> <reference anchor="Trammell-14"> <front> <title>Inline Data Integrity Signals for Passive Measurement, In: Dainotti A., Mahanti A., Uhlig S. (eds) Traffic Monitoring and Analysis. TMA 2014. Lecture Notes in Computer Science, vol 8406. Springer, Berlin, Heidelberg https://link.springer.com/chapter/10.1007/978-3-642-54999-1_2</title> <author fullname="B.Trammell, et al." initials="B." surname="Trammell"> <organization>TMA 2014 In: Dainotti A., Mahanti A., Uhlig S. (eds) Traffic Monitoring and Analysis. TMA 2014. Lecture Notes in Computer Science, vol 8406. Springer, Berlin, Heidelberg</organization> </author> <date month="March" year="2014"/> </front> </reference> </references> <references title="Informative References"> <?rfc include='reference.RFC.1242'?> <?rfc ?> <?rfc ?> <?rfc ?> <?rfc ?> <?rfc ?> <?rfc ?> <?rfc ?> <?rfc include='reference.RFC.6390'?> <?rfc include='reference.RFC.6703'?> <?rfc include='reference.RFC.7594'?> <?rfc ?> <?rfc ?> <?rfc ?> </references></back> </rfc>