Internet Engineering Task Force (IETF) N. Elkins
Request for Comments: 8250 Inside Products
Category: Standards Track R. Hamilton
ISSN: 2070-1721 Chemical Abstracts Service
M. Ackermann
BCBS Michigan
September 2017
IPv6 Performance and Diagnostic Metrics (PDM) Destination Option
Abstract
To assess performance problems, this document describes optional
headers embedded in each packet that provide sequence numbers and
timing information as a basis for measurements. Such measurements
may be interpreted in real time or after the fact. This document
specifies the Performance and Diagnostic Metrics (PDM) Destination
Options header. The field limits, calculations, and usage in
measurement of PDM are included in this document.
Status of This Memo
This is an Internet Standards Track document.
This document is a product of the Internet Engineering Task Force
(IETF). It represents the consensus of the IETF community. It has
received public review and has been approved for publication by the
Internet Engineering Steering Group (IESG). Further information on
Internet Standards is available in Section 2 of RFC 7841.
Information about the current status of this document, any errata,
and how to provide feedback on it may be obtained at
https://www.rfc-editor.org/info/rfc8250.
Copyright Notice
Copyright (c) 2017 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(https://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
Table of Contents
1. Background ......................................................3
1.1. Terminology ................................................3
1.2. Rationale for Defined Solution .............................4
1.3. IPv6 Transition Technologies ...............................4
2. Measurement Information Derived from PDM ........................5
2.1. Round-Trip Delay ...........................................5
2.2. Server Delay ...............................................5
3. Performance and Diagnostic Metrics Destination Option Layout ....6
3.1. Destination Options Header .................................6
3.2. Performance and Diagnostic Metrics Destination Option ......6
3.2.1. PDM Layout ..........................................6
3.2.2. Base Unit for Time Measurement ......................8
3.3. Header Placement ...........................................9
3.4. Header Placement Using IPsec ESP Mode ......................9
3.4.1. Using ESP Transport Mode ...........................10
3.4.2. Using ESP Tunnel Mode ..............................10
3.5. Implementation Considerations .............................10
3.5.1. PDM Activation .....................................10
3.5.2. PDM Timestamps .....................................10
3.6. Dynamic Configuration Options .............................11
3.7. Information Access and Storage ............................11
4. Security Considerations ........................................11
4.1. Resource Consumption and Resource Consumption Attacks .....11
4.2. Pervasive Monitoring ......................................12
4.3. PDM as a Covert Channel ...................................12
4.4. Timing Attacks ............................................13
5. IANA Considerations ............................................13
6. References .....................................................14
6.1. Normative References ......................................14
6.2. Informative References ....................................14
Appendix A. Context for PDM .......................................15
A.1. End-User Quality of Service (QoS) ..........................15
A.2. Need for a Packet Sequence Number (PSN) ....................15
A.3. Rationale for Defined Solution .............................15
A.4. Use PDM with Other Headers .................................16
Appendix B. Timing Considerations .................................16
B.1. Calculations of Time Differentials .........................16
B.2. Considerations of This Time-Differential Representation ....18
B.2.1. Limitations with This Encoding Method ..................18
B.2.2. Loss of Precision Induced by Timer Value Truncation ....19
Appendix C. Sample Packet Flows ...................................20
C.1. PDM Flow - Simple Client-Server Traffic ....................20
C.1.1. Step 1 .................................................20
C.1.2. Step 2 .................................................21
C.1.3. Step 3 .................................................21
C.1.4. Step 4 .................................................23
C.1.5. Step 5 .................................................24
C.2. Other Flows ................................................24
C.2.1. PDM Flow - One-Way Traffic .............................24
C.2.2. PDM Flow - Multiple-Send Traffic .......................25
C.2.3. PDM Flow - Multiple-Send Traffic with Errors ...........26
Appendix D. Potential Overhead Considerations .....................28
Acknowledgments ...................................................30
Authors' Addresses ................................................30
1. Background
To assess performance problems, measurements based on optional
sequence numbers and timing may be embedded in each packet. Such
measurements may be interpreted in real time or after the fact.
As defined in RFC 8200 [RFC8200], destination options are carried by
the IPv6 Destination Options extension header. Destination options
include optional information that need be examined only by the IPv6
node given as the destination address in the IPv6 header, not by
routers or other "middleboxes". This document specifies the
Performance and Diagnostic Metrics (PDM) destination option. The
field limits, calculations, and usage in measurement of the PDM
Destination Options header are included in this document.
1.1. Terminology
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 [RFC2119] [RFC8174] when, and only when, they appear in all
capitals, as shown here.
1.2. Rationale for Defined Solution
The current IPv6 specification does not provide timing, nor does it
provide a similar field in the IPv6 main header or in any extension
header. The IPv6 PDM destination option provides such fields.
Advantages include:
1. Real measure of actual transactions.
2. Ability to span organizational boundaries with consistent
instrumentation.
3. No time synchronization needed between session partners.
4. Ability to handle all transport protocols (TCP, UDP, the Stream
Control Transmission Protocol (SCTP), etc.) in a uniform way.
PDM provides the ability to determine quickly if the (latency)
problem is in the network or in the server (application). That is,
it is a fast way to do triage. For more information on the
background and usage of PDM, see Appendix A.
1.3. IPv6 Transition Technologies
In the path to full implementation of IPv6, transition technologies
such as translation or tunneling may be employed. It is possible
that an IPv6 packet containing PDM may be dropped if using IPv6
transition technologies. For example, an implementation using a
translation technique (IPv6 to IPv4) that does not support or
recognize the IPv6 Destination Options extension header may simply
drop the packet rather than translating it without the extension
header.
It is also possible that some devices in the network may not
correctly handle multiple IPv6 extension headers, including the IPv6
Destination Option. For example, adding the PDM header to a packet
may push the Layer 4 information to a point in the packet where it
is not visible to filtering logic, and the packet may be dropped.
This kind of situation is expected to become rare over time.
2. Measurement Information Derived from PDM
Each packet contains information about the sender and receiver. In
IP, the identifying information is called a "5-tuple".
The 5-tuple consists of:
SADDR: IP address of the sender
SPORT: Port for the sender
DADDR: IP address of the destination
DPORT: Port for the destination
PROTC: Upper-layer protocol (TCP, UDP, ICMP, etc.)
PDM contains the following base fields (scale fields intentionally
left out):
PSNTP : Packet Sequence Number This Packet
PSNLR : Packet Sequence Number Last Received
DELTATLR: Delta Time Last Received
DELTATLS: Delta Time Last Sent
Other fields for storing time scaling factors are also in PDM and
will be described in Section 3.
This information, combined with the 5-tuple, allows the measurement
of the following metrics:
1. Round-trip delay
2. Server delay
2.1. Round-Trip Delay
Round-trip *network* delay is the delay for packet transfer from a
source host to a destination host and then back to the source host.
This measurement has been defined, and its advantages and
disadvantages are discussed in "A Round-trip Delay Metric for IPPM"
[RFC2681].
2.2. Server Delay
Server delay is the interval between when a packet is received by a
device and the first corresponding packet is sent back in response.
This may be "server processing time". It may also be a delay caused
by acknowledgments. Server processing time includes the time taken
by the combination of the stack and application to return the
response. The stack delay may be related to network performance. If
this aggregate time is seen as a problem and there is a need to make
a clear distinction between application processing time and stack
delay, including that caused by the network, then more client-based
measurements are needed.
3. Performance and Diagnostic Metrics Destination Option Layout
3.1. Destination Options Header
The IPv6 Destination Options extension header [RFC8200] is used to
carry optional information that needs to be examined only by a
packet's destination node(s). The Destination Options header is
identified by a Next Header value of 60 in the immediately preceding
header and is defined in RFC 8200 [RFC8200]. The IPv6 Performance
and Diagnostic Metrics (PDM) destination option is implemented as an
IPv6 Option carried in the Destination Options header. PDM does not
require time synchronization.
3.2. Performance and Diagnostic Metrics Destination Option
3.2.1. PDM Layout
The IPv6 PDM destination option contains the following fields:
SCALEDTLR: Scale for Delta Time Last Received
SCALEDTLS: Scale for Delta Time Last Sent
PSNTP : Packet Sequence Number This Packet
PSNLR : Packet Sequence Number Last Received
DELTATLR : Delta Time Last Received
DELTATLS : Delta Time Last Sent
PDM has alignment requirements. Following the convention in IPv6,
these options are aligned in a packet so that multi-octet values
within the Option Data field of each option fall on natural
boundaries (i.e., fields of width n octets are placed at an integer
multiple of n octets from the start of the header, for n = 1, 2, 4,
or 8) [RFC8200].
The PDM destination option is encoded in type-length-value (TLV)
format as follows:
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Option Type | Option Length | ScaleDTLR | ScaleDTLS |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| PSN This Packet | PSN Last Received |
|-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Delta Time Last Received | Delta Time Last Sent |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Option Type
0x0F [RFC2780]
In keeping with RFC 8200 [RFC8200], the two high-order bits of
the Option Type field are encoded to indicate specific
processing of the option; for the PDM destination option, these
two bits MUST be set to 00.
The third high-order bit of the Option Type field specifies
whether or not the Option Data of that option can change
en route to the packet's final destination.
In PDM, the value of the third high-order bit MUST be 0.
Option Length
8-bit unsigned integer. Length of the option, in octets,
excluding the Option Type and Option Length fields. This field
MUST be set to 10.
Scale Delta Time Last Received (SCALEDTLR)
8-bit unsigned integer. This is the scaling value for the
Delta Time Last Received (DELTATLR) field. The possible values
are from 0 to 255. See Appendix B for further discussion on
timing considerations and formatting of the scaling values.
Scale Delta Time Last Sent (SCALEDTLS)
8-bit signed integer. This is the scaling value for the Delta
Time Last Sent (DELTATLS) field. The possible values are from
0 to 255.
Packet Sequence Number This Packet (PSNTP)
16-bit unsigned integer. This field will wrap. It is intended
for use while analyzing packet traces.
This field is initialized at a random number and incremented
monotonically for each packet of the session flow of the
5-tuple. The random-number initialization is intended to make
it harder to spoof and insert such packets.
Operating systems MUST implement a separate packet sequence
number counter per 5-tuple.
Packet Sequence Number Last Received (PSNLR)
16-bit unsigned integer. This is the PSNTP of the packet last
received on the 5-tuple.
This field is initialized to 0.
Delta Time Last Received (DELTATLR)
16-bit unsigned integer. The value is set according to the
scale in SCALEDTLR.
Delta Time Last Received =
(send time packet n - receive time packet (n - 1))
Delta Time Last Sent (DELTATLS)
16-bit unsigned integer. The value is set according to the
scale in SCALEDTLS.
Delta Time Last Sent =
(receive time packet n - send time packet (n - 1))
3.2.2. Base Unit for Time Measurement
A time differential is always a whole number in a CPU; it is the unit
specification -- hours, seconds, nanoseconds -- that determines what
the numeric value means. For PDM, the base time unit is 1 attosecond
(asec). This allows for a common unit and scaling of the time
differential among all IP stacks and hardware implementations.
Note that PDM provides the ability to measure both time differentials
that are extremely small and time differentials in a Delay/Disruption
Tolerant Networking (DTN) environment where the delays may be very
great. To store a time differential in just 16 bits that must range
in this way will require some scaling of the time-differential value.
One issue is the conversion from the native time base in the CPU
hardware of whatever device is in use to some number of attoseconds.
It might seem that this will be an astronomical number, but the
conversion is straightforward. It involves multiplication by an
appropriate power of 10 to change the value into a number of
attoseconds. Then, to scale the value so that it fits into DELTATLR
or DELTATLS, the value is shifted by a number of bits, retaining the
16 high-order or most significant bits. The number of bits shifted
becomes the scaling factor, stored as SCALEDTLR or SCALEDTLS,
respectively. For additional information on this process, see
Appendix B.
3.3. Header Placement
The PDM destination option is placed as defined in RFC 8200
[RFC8200]. There may be a choice of where to place the Destination
Options header. If using Encapsulating Security Payload (ESP) mode,
please see Section 3.4 of this document regarding the placement of
the PDM Destination Options header.
For each IPv6 packet header, PDM MUST NOT appear more than once.
However, an encapsulated packet MAY contain a separate PDM associated
with each encapsulated IPv6 header.
3.4. Header Placement Using IPsec ESP Mode
IPsec ESP is defined in [RFC4303] and is widely used. Section 3.1.1
of [RFC4303] discusses the placement of Destination Options headers.
The placement of PDM is different, depending on whether ESP is used
in tunnel mode or transport mode.
In the ESP case, no 5-tuple is available, as there are no port
numbers. ESP flow should be identified only by using SADDR, DADDR,
and PROTC. The Security Parameter Index (SPI) numbers SHOULD be
ignored when considering the flow over which PDM information is
measured.
3.4.1. Using ESP Transport Mode
Note that destination options may be placed before or after ESP, or
both. If using PDM in ESP transport mode, PDM MUST be placed after
the ESP header so as not to leak information.
3.4.2. Using ESP Tunnel Mode
Note that in both the outer set of IP headers and the inner set of IP
headers, destination options may be placed before or after ESP, or
both. A tunnel endpoint that creates a new packet may decide to use
PDM independently of the use of PDM of the original packet to enable
delay measurements between the two tunnel endpoints.
3.5. Implementation Considerations
3.5.1. PDM Activation
An implementation should provide an interface to enable or disable
the use of PDM. This specification recommends having PDM off by
default.
PDM MUST NOT be turned on merely if a packet is received with a PDM
header. The received packet could be spoofed by another device.
3.5.2. PDM Timestamps
The PDM timestamps are intended to isolate wire time from server or
host time but may necessarily attribute some host processing time to
network latency.
Section 10.2 of RFC 2330 [RFC2330] ("Framework for IP Performance
Metrics") describes two notions of "wire time". These notions are
only defined in terms of an Internet host H observing an Internet
link L at a particular location:
+ For a given IP packet P, the "wire arrival time" of P at H on L is
the first time T at which any bit of P has appeared at H's
observational position on L.
+ For a given IP packet P, the "wire exit time" of P at H on L is
the first time T at which all the bits of P have appeared at H's
observational position on L.
This specification does not define H's exact observational position
on L. That is left for the deployment setups to define. However,
the position where PDM timestamps are taken SHOULD be as close to the
physical network interface as possible. Not all implementations will
be able to achieve the ideal level of measurement.
3.6. Dynamic Configuration Options
If the PDM Destination Options header is used, then it MAY be turned
on for all packets flowing through the host, applied to an upper-
layer protocol (TCP, UDP, SCTP, etc.), a local port, or IP address
only. These are at the discretion of the implementation.
3.7. Information Access and Storage
Measurement information provided by PDM may be made accessible for
higher layers or the user itself. Similar to activating the use of
PDM, the implementation may also provide an interface to indicate if
received.
PDM information may be stored, if desired. If a packet with PDM
information is received and the information should be stored, the
upper layers may be notified. Furthermore, the implementation should
define a configurable maximum lifetime after which the information
can be removed as well as a configurable maximum amount of memory
that should be allocated for PDM information.
4. Security Considerations
PDM may introduce some new security weaknesses.
4.1. Resource Consumption and Resource Consumption Attacks
PDM needs to calculate the deltas for time and keep track of the
sequence numbers. This means that control blocks that reside in
memory may be kept at the end hosts per 5-tuple.
A limit on how much memory is being used SHOULD be implemented.
Without a memory limit, any time that a control block is kept in
memory, an attacker can try to misuse the control blocks to cause
excessive resource consumption. This could be used to compromise the
end host.
PDM is used only at the end hosts, and memory is used only at the end
host and not at routers or middleboxes.
4.2. Pervasive Monitoring
Since PDM passes in the clear, a concern arises as to whether the
data can be used to fingerprint the system or somehow obtain
information about the contents of the payload.
Let us discuss fingerprinting of the end host first. It is possible
that seeing the pattern of deltas or the absolute values could give
some information as to the speed of the end host -- that is, if it is
a very fast system or an older, slow device. This may be useful to
the attacker. However, if the attacker has access to PDM, the
attacker also has access to the entire packet and could make such a
deduction based merely on the time frames elapsed between packets
WITHOUT PDM.
As far as deducing the content of the payload, in terms of the
application-level information such as web page, user name, user
password, and so on, it appears to us that PDM is quite unhelpful in
this regard. Having said that, the ability to separate wire time
from processing time may potentially provide an attacker with
additional information. It is conceivable that an attacker could
attempt to deduce the type of application in use by noting the server
time and payload length. Some encryption algorithms attempt to
obfuscate the packet length to avoid just such vulnerabilities. In
the future, encryption algorithms may wish to obfuscate the server
time as well.
4.3. PDM as a Covert Channel
PDM provides a set of fields in the packet that could be used to leak
data. But there is no real reason to suspect that PDM would be
chosen rather than another part of the payload or another extension
header.
A firewall or another device could sanity-check the fields within
PDM, but randomly assigned sequence numbers and delta times might be
expected to vary widely. The biggest problem, though, is how an
attacker would get access to PDM in the first place to leak data.
The attacker would have to either compromise the end host or have a
Man in the Middle (MitM). It is possible that either one could
change the fields, but the other end host would then get sequence
numbers and deltas that don't make any sense.
It is conceivable that someone could compromise an end host and make
it start sending packets with PDM without the knowledge of the host.
But, again, the bigger problem is the compromise of the end host.
Once that is done, the attacker probably has better ways to
leak data.
Having said that, if a PDM-aware middlebox or an implementation
(destination host) detects some number of "nonsensical" sequence
numbers or timing information, it could take action to block this
traffic, discard it, or send an alert.
4.4. Timing Attacks
The fact that PDM can help in the separation of node processing time
from network latency brings value to performance monitoring. Yet, it
is this very characteristic of PDM that may be misused to make
certain new types of timing attacks against protocols and
implementations possible.
Depending on the nature of the cryptographic protocol used, it may be
possible to leak the credentials of the device. For example, if an
attacker can see that PDM is being used, then the attacker might use
PDM to launch a timing attack against the keying material used by the
cryptographic protocol.
An implementation may want to be sure that PDM is enabled only for
certain IP addresses or only for some ports. Additionally, the
implementation SHOULD require an explicit restart of monitoring after
a certain time period (for example, after 1 hour) to make sure that
PDM is not accidentally left on (for example, after debugging has
been done).
Even so, if using PDM, a user "Consent to be Measured" SHOULD be a
prerequisite for using PDM. Consent is common in enterprises and
with some subscription services. The actual content of "Consent to
be Measured" will differ by site, but it SHOULD make clear that the
traffic is being measured for Quality of Service (QoS) and to assist
in diagnostics, as well as to make clear that there may be potential
risks of certain vulnerabilities if the traffic is captured during a
diagnostic session.
5. IANA Considerations
IANA has registered a Destination Option Type assignment with the act
bits set to 00 and the chg bit set to 0 from the "Destination Options
and Hop-by-Hop Options" sub-registry of the "Internet Protocol
Version 6 (IPv6) Parameters" registry [RFC2780] at
<https://www.iana.org/assignments/ipv6-parameters/>.
Hex Value Binary Value Description Reference
act chg rest
---------------------------------------------------------------------
0x0F 00 0 01111 Performance and RFC 8250
Diagnostic Metrics (PDM)
6. References
6.1. Normative References
[RFC1122] Braden, R., Ed., "Requirements for Internet Hosts -
Communication Layers", STD 3, RFC 1122,
DOI 10.17487/RFC1122, October 1989,
<https://www.rfc-editor.org/info/rfc1122>.
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>.
[RFC2681] Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip
Delay Metric for IPPM", RFC 2681, DOI 10.17487/RFC2681,
September 1999, <https://www.rfc-editor.org/info/rfc2681>.
[RFC2780] Bradner, S. and V. Paxson, "IANA Allocation Guidelines For
Values In the Internet Protocol and Related Headers",
BCP 37, RFC 2780, DOI 10.17487/RFC2780, March 2000,
<https://www.rfc-editor.org/info/rfc2780>.
[RFC4303] Kent, S., "IP Encapsulating Security Payload (ESP)",
RFC 4303, DOI 10.17487/RFC4303, December 2005,
<https://www.rfc-editor.org/info/rfc4303>.
[RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in
RFC 2119 Key Words", BCP 14, RFC 8174,
DOI 10.17487/RFC8174, May 2017,
<https://www.rfc-editor.org/info/rfc8174>.
[RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6
(IPv6) Specification", STD 86, RFC 8200,
DOI 10.17487/RFC8200, July 2017,
<https://www.rfc-editor.org/info/rfc8200>.
6.2. Informative References
[RFC2330] Paxson, V., Almes, G., Mahdavi, J., and M. Mathis,
"Framework for IP Performance Metrics", RFC 2330,
DOI 10.17487/RFC2330, May 1998,
<https://www.rfc-editor.org/info/rfc2330>.
[TCPM] Scheffenegger, R., Kuehlewind, M., and B. Trammell,
"Encoding of Time Intervals for the TCP Timestamp Option",
Work in Progress, draft-trammell-tcpm-timestamp-
interval-01, July 2013.
Appendix A. Context for PDM
A.1. End-User Quality of Service (QoS)
The timing values in PDM embedded in the packet will be used to
estimate QoS as experienced by an end-user device.
For many applications, the key user performance indicator is response
time. When the end user is an individual, he is generally
indifferent to what is happening along the network; what he really
cares about is how long it takes to get a response back. But this is
not just a matter of individuals' personal convenience. In many
cases, rapid response is critical to the business being conducted.
Low, reliable, and acceptable response times are not just "nice to
have". On many networks, the impact can be financial hardship or can
endanger human life. In some cities, the emergency police contact
system operates over IP; all levels of law enforcement use IP
networks; transactions on our stock exchanges are settled using IP
networks. The critical nature of such activities to our daily lives
and financial well-being demands a simple solution to support
response-time measurements.
A.2. Need for a Packet Sequence Number (PSN)
While performing network diagnostics on an end-to-end connection, it
often becomes necessary to isolate the factors along the network path
responsible for problems. Diagnostic data may be collected at
multiple places along the path (if possible), or at the source and
destination. Then, in post-collection processing, the diagnostic
data corresponding to each packet at different observation points
must be matched for proper measurements. A sequence number in each
packet provides a sufficient basis for the matching process. If
need be, the timing fields may be used along with the sequence number
to ensure uniqueness.
This method of data collection along the path is of special use for
determining where packet loss or packet corruption is happening.
The packet sequence number needs to be unique in the context of the
session (5-tuple).
A.3. Rationale for Defined Solution
One of the important functions of PDM is to allow you to quickly
dispatch the right set of diagnosticians. Within network or server
latency, there may be many components. The job of the diagnostician
is to rule each one out until the culprit is found.
PDM will fit into this diagnostic picture by quickly telling you how
to escalate. PDM will point to either the network area or the server
area. Within the server latency, PDM does not tell you whether the
bottleneck is in the IP stack, the application, or buffer allocation.
Within the network latency, PDM does not tell you which of the
network segments or middleboxes is at fault.
What PDM does tell you is whether the problem is in the network or
the server.
A.4. Use PDM with Other Headers
For diagnostics, one may want to use PDM with other headers (Layer 2,
Layer 3, etc). For example, if PDM is used by a technician (or tool)
looking at a packet capture, within the packet capture, they would
have available to them the Layer 2 header, IP header (v6 or v4), TCP
header, UDP header, ICMP header, SCTP header, or other headers. All
information would be looked at together to make sense of the packet
flow. The technician or processing tool could analyze, report, or
ignore the data from PDM, as necessary.
For an example of how PDM can help with TCP retransmission problems,
please look at Appendix C.
Appendix B. Timing Considerations
B.1. Calculations of Time Differentials
When SCALEDTLR or SCALEDTLS is used, it means that the description of
the processing applies equally to SCALEDTLR and SCALEDTLS.
The time counter in a CPU is a binary whole number representing a
number of milliseconds (msec), microseconds (usec), or even
picoseconds (psec). Representing one of these values as attoseconds
(asec) means multiplying by 10 raised to some exponent. Refer to
this table of equalities:
Base value = # of sec = # of asec 1000s of asec
--------------- ------------- ------------- -------------
1 second 1 sec 10**18 asec 1000**6 asec
1 millisecond 10**-3 sec 10**15 asec 1000**5 asec
1 microsecond 10**-6 sec 10**12 asec 1000**4 asec
1 nanosecond 10**-9 sec 10**9 asec 1000**3 asec
1 picosecond 10**-12 sec 10**6 asec 1000**2 asec
1 femtosecond 10**-15 sec 10**3 asec 1000**1 asec
For example, if you have a time differential expressed in
microseconds, since each microsecond is 10**12 asec, you would
multiply your time value by 10**12 to obtain the number of
attoseconds. If your time differential is expressed in nanoseconds,
you would multiply by 10**9 to get the number of attoseconds.
The result is a binary value that will need to be shortened by a
number of bits so it will fit into the 16-bit PDM delta field.
The next step is to divide by 2 until the value is contained in just
16 significant bits. The exponent of the value in the last column of
the table is useful here; the initial scaling factor is that exponent
multiplied by 10. This is the minimum number of low-order bits to be
shifted out or discarded. It represents dividing the time value by
1024 raised to that exponent.
The resulting value may still be too large to fit into 16 bits but
can be normalized by shifting out more bits (dividing by 2) until the
value fits into the 16-bit delta field. The number of extra bits
shifted out is then added to the scaling factor. The scaling factor
-- the total number of low-order bits dropped -- is the SCALEDTLR or
SCALEDTLS value.
For example, say an application has these start and finish timer
values (hexadecimal values, in microseconds):
Finish: 27C849234 usec (02:57:58.997556)
-Start: 27C83F696 usec (02:57:58.957718)
========== ============== ==========================
Difference 9B9E usec 0.039838 sec or 39838 usec
To convert this differential value to binary attoseconds, multiply
the number of microseconds by 10**12. Divide by 1024**4, or simply
discard 40 bits from the right. The result is 36232, or 8D88 in hex,
with a scaling factor or SCALEDTLR/SCALEDTLS value of 40.
For another example, presume the time differential is larger, say
32.311072 seconds, which is 32311072 usec. Each microsecond is
10**12 asec, so multiply by 10**12, giving the hexadecimal value
1C067FCCAE8120000. Using the initial scaling factor of 40, drop the
last 10 characters (40 bits) from that string, giving 1C067FC. This
will not fit into a delta field, as it is 25 bits long. Shifting the
value to the right another 9 bits results in a delta value of E033,
with a resulting scaling factor of 49.
When the time-differential value is a small number, regardless of the
time unit, the exponent trick given above is not useful in
determining the proper scaling value. For example, if the time
differential is 3 seconds and you want to convert that directly, you
would follow this path:
3 seconds = 3*10**18 asec (decimal)
= 29A2241AF62C0000 asec (hexadecimal)
If you just truncate the last 60 bits, you end up with a delta value
of 2 and a scaling factor of 60, when what you really wanted was a
delta value with more significant digits. The most precision with
which you can store this value in 16 bits is A688, with a scaling
factor of 46.
B.2. Considerations of This Time-Differential Representation
There are two considerations to be taken into account with this
representation of a time differential. The first is whether there
are any limitations on the maximum or minimum time differential that
can be expressed using the method of a delta value and a scaling
factor. The second is the amount of imprecision introduced by this
method.
B.2.1. Limitations with This Encoding Method
The DELTATLS and DELTATLR fields store only the 16 most significant
bits of the time-differential value. Thus, the range, excluding the
scaling factor, is from 0 to 65535, or a maximum of 2**16 - 1. This
method is further described in [TCPM].
The actual magnitude of the time differential is determined by the
scaling factor. SCALEDTLR and SCALEDTLS are 8-bit unsigned integers,
so the scaling factor ranges from 0 to 255. The smallest number that
can be represented would have a value of 1 in the delta field and a
value of 0 in the associated scale field. This is the representation
for 1 attosecond. Clearly, this allows PDM to measure extremely
small time differentials.
On the other end of the scale, the maximum delta value is 65535, or
FFFF in hexadecimal. If the maximum scale value of 255 is used, the
time differential represented is 65535*2**255, which is over
3*10**55 years -- essentially, forever. So, there appears to be no
real limitation to the time differential that can be represented.
B.2.2. Loss of Precision Induced by Timer Value Truncation
As PDM specifies the DELTATLR and DELTATLS values as 16-bit unsigned
integers, any time that the precision is greater than those 16 bits,
there will be truncation of the trailing bits, with an accompanying
loss of precision in the value.
Any time-differential value smaller than 65536 asec can be stored
exactly in DELTATLR or DELTATLS, because the representation of this
value requires at most 16 bits.
Since the time-differential values in PDM are measured in
attoseconds, the range of values that would be truncated to the same
encoded value is 2**((Scale) - 1) asec.
For example, the smallest time differential that would be truncated
to fit into a delta field is
1 0000 0000 0000 0000 = 65536 asec
This value would be encoded as a delta value of 8000 (hexadecimal)
with a scaling factor of 1. The value
1 0000 0000 0000 0001 = 65537 asec
would also be encoded as a delta value of 8000 with a scaling factor
of 1. This actually is the largest value that would be truncated to
that same encoded value. When the scale value is 1, the value range
is calculated as 2**1 - 1, or 1 asec, which you can see is the
difference between these minimum and maximum values.
The scaling factor is defined as the number of low-order bits
truncated to reduce the size of the resulting value so it fits into a
16-bit delta field. If, for example, you had to truncate 12 bits,
the loss of precision would depend on the bits you truncated. The
range of these values would be
0000 0000 0000 = 0 asec
to
1111 1111 1111 = 4095 asec
So, the minimum loss of precision would be 0 asec, where the delta
value exactly represents the time differential, and the maximum loss
of precision would be 4095 asec. As stated above, the scaling factor
of 12 means that the maximum loss of precision is 2**12 - 1 asec, or
4095 asec.
Compare this loss of precision to the actual time differential. The
range of actual time-differential values that would incur this loss
of precision is from
1000 0000 0000 0000 0000 0000 0000 = 2**27 asec or 134217728 asec
to
1111 1111 1111 1111 1111 1111 1111 = 2**28 - 1 asec or 268435455 asec
Granted, these are small values, but the point is that any value
between these two values will have a maximum loss of precision of
4095 asec, or about 0.00305% for the first value, as encoded, and at
most 0.001526% for the second. These maximum-loss percentages are
consistent for all scaling values.
Appendix C. Sample Packet Flows
C.1. PDM Flow - Simple Client-Server Traffic
Below is a sample simple flow for PDM with one packet sent from
Host A and one packet received by Host B. PDM does not require time
synchronization between Host A and Host B. The calculations to
derive meaningful metrics for network diagnostics are shown below
each packet sent or received.
C.1.1. Step 1
Packet 1 is sent from Host A to Host B. The time for Host A is set
initially to 10:00AM.
The time and packet sequence number are saved by the sender
internally. The packet sequence number and delta times are sent in
the packet.
Packet 1
+----------+ +----------+
| | | |
| Host | ----------> | Host |
| A | | B |
| | | |
+----------+ +----------+
PDM Contents:
PSNTP : Packet Sequence Number This Packet: 25
PSNLR : Packet Sequence Number Last Received: -
DELTATLR : Delta Time Last Received: -
SCALEDTLR: Scale of Delta Time Last Received: 0
DELTATLS : Delta Time Last Sent: -
SCALEDTLS: Scale of Delta Time Last Sent: 0
Internally, within the sender, Host A, it must keep:
Packet Sequence Number of the last packet sent: 25
Time the last packet was sent: 10:00:00
Note: The initial PSNTP from Host A starts at a random number -- in
this case, 25. The time in these examples is shown in seconds for
the sake of simplicity.
C.1.2. Step 2
Packet 1 is received at Host B. Its time is set to 1 hour later than
Host A -- in this case, 11:00AM.
Internally, within the receiver, Host B, it must note the following:
Packet Sequence Number of the last packet received: 25
Time the last packet was received : 11:00:03
Note: This timestamp is in Host B time. It has nothing whatsoever to
do with Host A time. The packet sequence number of the last packet
received will become PSNLR, which will be sent out in the packet sent
by Host B in the next step. The timestamp of the packet last
received (as noted above) will be used as input to calculate the
DELTATLR value to be sent out in the packet sent by Host B in the
next step.
C.1.3. Step 3
Packet 2 is sent by Host B to Host A. Note that the initial packet
sequence number (PSNTP) from Host B starts at a random number -- in
this case, 12. Before sending the packet, Host B does a calculation
of deltas. Since Host B knows when it is sending the packet and it
knows when it received the previous packet, it can do the following
calculation:
DELTATLR = send time (packet 2) - receive time (packet 1)
Note: Both the send time and the receive time are saved internally in
Host B. They do not travel in the packet. Only the change in values
(delta) is in the packet. This is the DELTATLR value.
Assume that within Host B we have the following:
Packet Sequence Number of the last packet received: 25
Time the last packet was received: 11:00:03
Packet Sequence Number of this packet: 12
Time this packet is being sent: 11:00:07
A delta value to be sent out in the packet can now be calculated.
DELTATLR becomes:
4 seconds = 11:00:07 - 11:00:03 = 3782DACE9D900000 asec
This is the derived metric: server delay. The time scaling factors
must be converted; in this case, the time differential is DE0B, and
the scaling factor is 2E, or 46 in decimal. Then, these values,
along with the packet sequence numbers, will be sent to Host A as
follows:
Packet 2
+----------+ +----------+
| | | |
| Host | <---------- | Host |
| A | | B |
| | | |
+----------+ +----------+
PDM Contents:
PSNTP : Packet Sequence Number This Packet: 12
PSNLR : Packet Sequence Number Last Received: 25
DELTATLR : Delta Time Last Received: DE0B (4 seconds)
SCALEDTLR: Scale of Delta Time Last Received: 2E (46 decimal)
DELTATLS : Delta Time Last Sent: -
SCALEDTLS: Scale of Delta Time Last Sent: 0
The metric left to be calculated is the round-trip delay. This will
be calculated by Host A when it receives packet 2.
C.1.4. Step 4
Packet 2 is received at Host A. Remember that its time is set to
1 hour earlier than Host B. Internally, it must note the following:
Packet Sequence Number of the last packet received: 12
Time the last packet was received : 10:00:12
Note: This timestamp is in Host A time. It has nothing whatsoever to
do with Host B time.
So, Host A can now calculate total end-to-end time. That is:
End-to-End Time = Time Last Received - Time Last Sent
For example, packet 25 was sent by Host A at 10:00:00. Packet 12 was
received by Host A at 10:00:12, so:
End-to-End time = 10:00:12 - 10:00:00 or 12 (server and network
round-trip delay combined).
This time may also be called "total overall Round-Trip Time
(RTT)", which includes network RTT and host response time.
We will call this derived metric "Delta Time Last Sent" (DELTATLS).
Round-trip delay can now be calculated. The formula is:
Round-trip delay =
(Delta Time Last Sent - Delta Time Last Received)
Or:
Round-trip delay = 12 - 4 or 8
At this point, the only problem is that all metrics are in Host A
only and not exposed in a packet. To do that, we need a third
packet.
Note: This simple example assumes one send and one receive. That is
done only for purposes of explaining the function of PDM. In cases
where there are multiple packets returned, one would take the time in
the last packet in the sequence. The calculations of such timings
and intelligent processing are the function of post-processing of
the data.
C.1.5. Step 5
Packet 3 is sent from Host A to Host B.
+----------+ +----------+
| | | |
| Host | ----------> | Host |
| A | | B |
| | | |
+----------+ +----------+
PDM Contents:
PSNTP : Packet Sequence Number This Packet: 26
PSNLR : Packet Sequence Number Last Received: 12
DELTATLR : Delta Time Last Received: 0
SCALEDTLS: Scale of Delta Time Last Received 0
DELTATLS : Delta Time Last Sent: A688 (scaled value)
SCALEDTLR: Scale of Delta Time Last Received: 30 (48 decimal)
To calculate two-way delay, any packet-capture device may look at
these packets and do what is necessary.
C.2. Other Flows
What has been discussed so far is a simple flow with one packet sent
and one returned. Let's look at how PDM may be useful in other types
of flows.
C.2.1. PDM Flow - One-Way Traffic
The flow on a particular session may not be a send-receive paradigm.
Let us consider some other situations. In the case of a one-way
flow, one might see the following.
Note: The time is expressed in generic units for simplicity. That
is, these values do not represent a number of attoseconds,
microseconds, or any other real units of time.
Packet Sender PSN PSN Delta Time Delta Time
This Packet Last Recvd Last Recvd Last Sent
=====================================================================
1 Server 1 0 0 0
2 Server 2 0 0 5
3 Server 3 0 0 12
4 Server 4 0 0 20
What does this mean, and how is it useful?
In a one-way flow, only the Delta Time Last Sent will be seen as
used. Recall that Delta Time Last Sent is the difference between the
send of one packet from a device and the next. This is a measure of
throughput for the sender -- according to the sender's point of view.
That is, it is a measure of how fast the application itself (with
stack time included) is able to send packets.
How might this be useful? If one is having a performance issue at
the client and sees that packet 2, for example, is sent after 5 time
units from the server but takes 10 times that long to arrive at the
destination, then one may safely conclude that there are delays in
the path, other than at the server, that may be causing the delivery
issue for that packet. Such delays may include the network links,
middleboxes, etc.
True one-way traffic is quite rare. What people often mean by
"one-way" traffic is an application such as FTP where a group of
packets (for example, a TCP window size worth) is sent and the sender
then waits for acknowledgment. This type of flow would actually fall
into the "multiple-send" traffic model.
C.2.2. PDM Flow - Multiple-Send Traffic
Assume that two packets are sent from the server and then an ACK is
sent from the client. For example, a TCP flow will do this, per
RFC 1122 [RFC1122], Section 4.2.3. Packets 1 and 2 are sent from the
server, and then an ACK is sent from the client. Packet 4 starts a
second sequence from the server.
Packet Sender PSN PSN Delta Time Delta Time
This Packet Last Recvd Last Recvd Last Sent
=====================================================================
1 Server 1 0 0 0
2 Server 2 0 0 5
3 Client 1 2 20 0
4 Server 3 1 10 15
How might this be used?
Notice that in packet 3, the client has a Delta Time Last Received
value of 20. Recall that:
DELTATLR = send time (packet 3) - receive time (packet 2)
So, what does one know now? In this case, Delta Time Last Received
is the processing time for the client to send the next packet.
How to interpret this depends on what is actually being sent.
Remember that PDM is not being used in isolation; rather, it is used
to supplement the fields found in other headers. Let's take two
examples:
1. The client is sending a standalone TCP ACK. One would find this
by looking at the payload length in the IPv6 header and the TCP
Acknowledgment field in the TCP header. So, in this case, the
client is taking 20 time units to send back the ACK. This may or
may not be interesting.
2. The client is sending data with the packet. Again, one would find
this by looking at the payload length in the IPv6 header and the
TCP Acknowledgment field in the TCP header. So, in this case, the
client is taking 20 time units to send back data. This may
represent "User Think Time". Again, this may or may not be
interesting in isolation. But if there is a performance problem
receiving data at the server, then, taken in conjunction with RTT
or other packet timing information, this information may be quite
interesting.
Of course, one also needs to look at the PSN Last Received field to
make sure of the interpretation of this data -- that is, to make sure
that the Delta Time Last Received corresponds to the packet of
interest.
The benefits of PDM are that such information is now available in a
uniform manner for all applications and all protocols without
extensive changes required to applications.
C.2.3. PDM Flow - Multiple-Send Traffic with Errors
Let us now look at a case of how PDM may be able to help in a case of
TCP retransmission and add to the information that is sent in the TCP
header.
Assume that three packets are sent with each send from the server.
From the server, this is what is seen:
Pkt Sender PSN PSN Delta Time Delta Time TCP Data
This Pkt Last Recvd Last Recvd Last Sent SEQ Bytes
=====================================================================
1 Server 1 0 0 0 123 100
2 Server 2 0 0 5 223 100
3 Server 3 0 0 5 333 100
The client, however, does not receive all the packets. From the
client, this is what is seen for the packets sent from the server:
Pkt Sender PSN PSN Delta Time Delta Time TCP Data
This Pkt Last Recvd Last Recvd Last Sent SEQ Bytes
=====================================================================
1 Server 1 0 0 0 123 100
2 Server 3 0 0 5 333 100
Let's assume that the server now retransmits the packet. (Obviously,
a duplicate acknowledgment sequence for fast retransmit or a
retransmit timeout would occur. To illustrate the point, these
packets are being left out.)
So, if a TCP retransmission is done, then from the client, this is
what is seen for the packets sent from the server:
Pkt Sender PSN PSN Delta Time Delta Time TCP Data
This Pkt Last Recvd Last Recvd Last Sent SEQ Bytes
=====================================================================
1 Server 4 0 0 30 223 100
The server has resent the old packet 2 with a TCP sequence number
of 223. The retransmitted packet now has a PSN This Packet
value of 4.
The Delta Time Last Sent is 30 -- in other words, the time between
sending the packet with a PSN of 3 and this current packet.
Let's say that packet 4 is lost again. Then, after some amount of
time (RTO), the packet with a TCP sequence number of 223 is resent.
From the client, this is what is seen for the packets sent from the
server:
Pkt Sender PSN PSN Delta Time Delta Time TCP Data
This Pkt Last Recvd Last Recvd Last Sent SEQ Bytes
====================================================================
1 Server 5 0 0 60 223 100
If this packet now arrives at the destination, one has a very good
idea that packets exist that are being sent from the server as
retransmissions and not arriving at the client. This is because the
PSN of the resent packet from the server is 5 rather than 4. If we
had used the TCP sequence number alone, we would never have seen this
situation. The TCP sequence number in all situations is 223.
This situation would be experienced by the user of the application
(the human being actually sitting somewhere) as "hangs" or long
delays between packets. On large networks, to diagnose problems such
as these where packets are lost somewhere on the network, one has to
take multiple traces to find out exactly where.
The first thing to do is to start with doing a trace at the client
and the server, so that we can see if the server sent a particular
packet and the client received it. If the client did not receive it,
then we start tracking back to trace points at the router right after
the server and the router right before the client. Did they get
these packets that the server has sent? This is a time-consuming
activity.
With PDM, we can speed up the diagnostic time because we may be able
to use only the trace taken at the client to see what the server is
sending.
Appendix D. Potential Overhead Considerations
One might wonder about the potential overhead of PDM. First, PDM is
entirely optional. That is, a site may choose to implement PDM or
not, as they wish. If they are happy with the costs of PDM versus
the benefits, then the choice should be theirs.
Below is a table outlining the potential overhead in terms of
additional time to deliver the response to the end user for various
assumed RTTs:
Bytes RTT Bytes Bytes New Overhead
in Packet Per Millisec in PDM RTT
====================================================================
1000 1000 milli 1 16 1016.000 16.000 milli
1000 100 milli 10 16 101.600 1.600 milli
1000 10 milli 100 16 10.160 0.160 milli
1000 1 milli 1000 16 1.016 0.016 milli
Below are two examples of actual RTTs for packets traversing large
enterprise networks.
The first example is for packets going to multiple business partners:
Bytes RTT Bytes Bytes New Overhead
in Packet Per Millisec in PDM RTT
====================================================================
1000 17 milli 58 16 17.360 0.360 milli
The second example is for packets at a large enterprise customer
within a data center. Notice that the scale is now in microseconds
rather than milliseconds:
Bytes RTT Bytes Bytes New Overhead
in Packet Per Microsec in PDM RTT
====================================================================
1000 20 micro 50 16 20.320 0.320 micro
As with other diagnostic tools, such as packet traces, a certain
amount of processing time will be required to create and process PDM.
Since PDM is lightweight (has only a few variables), we expect the
processing time to be minimal.
Acknowledgments
The authors would like to thank Keven Haining, Al Morton, Brian
Trammell, David Boyes, Bill Jouris, Richard Scheffenegger, and Rick
Troth for their comments and assistance. We would also like to thank
Tero Kivinen and Jouni Korhonen for their detailed and perceptive
reviews.
Authors' Addresses
Nalini Elkins
Inside Products, Inc.
36A Upper Circle
Carmel Valley, CA 93924
United States of America
Phone: +1 831 659 8360
Email: nalini.elkins@insidethestack.com
URI: http://www.insidethestack.com
Robert M. Hamilton
Chemical Abstracts Service
A Division of the American Chemical Society
2540 Olentangy River Road
Columbus, Ohio 43202
United States of America
Phone: +1 614 447 3600 x2517
Email: rhamilton@cas.org
URI: http://www.cas.org
Michael S. Ackermann
Blue Cross Blue Shield of Michigan
P.O. Box 2888
Detroit, Michigan 48231
United States of America
Phone: +1 310 460 4080
Email: mackermann@bcbsm.com
URI: http://www.bcbsm.com