Debian Codegraph @

rfc4689.html

raw
1 <pre>Network Working Group S. Poretsky
2 Request for Comments: 4689 Reef Point Systems
3 Category: Informational J. Perser
4 Veriwave
5 S. Erramilli
6 Telcordia
7 S. Khurana
8 Motorola
9 October 2006
10
11
12 <span class="h1">Terminology for Benchmarking Network-layer Traffic Control Mechanisms</span>
13
14 Status of This Memo
15
16 This memo provides information for the Internet community. It does
17 not specify an Internet standard of any kind. Distribution of this
18 memo is unlimited.
19
20 Copyright Notice
21
22 Copyright (C) The Internet Society (2006).
23
24 Abstract
25
26 This document describes terminology for the benchmarking of devices
27 that implement traffic control using packet classification based on
28 defined criteria. The terminology is to be applied to measurements
29 made on the data plane to evaluate IP traffic control mechanisms.
30 Rules for packet classification can be based on any field in the IP
31 header, such as the Differentiated Services Code Point (DSCP), or any
32 field in the packet payload, such as port number.
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52 <span class="grey">Poretsky, et al. Informational [Page 1]</span></pre>
53 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-2" ></span>
54 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
55
56
57 Table of Contents
58
59 <a href="#section-1">1</a>. Introduction ....................................................<a href="#page-2">2</a>
60 <a href="#section-2">2</a>. Existing Definitions ............................................<a href="#page-3">3</a>
61 <a href="#section-3">3</a>. Term Definitions ................................................<a href="#page-4">4</a>
62 <a href="#section-3.1">3.1</a>. Configuration Terms ........................................<a href="#page-4">4</a>
63 <a href="#section-3.1.1">3.1.1</a>. Classification ......................................<a href="#page-4">4</a>
64 <a href="#section-3.1.2">3.1.2</a>. Codepoint Set .......................................<a href="#page-4">4</a>
65 <a href="#section-3.1.3">3.1.3</a>. Forwarding Congestion ...............................<a href="#page-5">5</a>
66 <a href="#section-3.1.4">3.1.4</a>. Congestion Management ...............................<a href="#page-6">6</a>
67 <a href="#section-3.1.5">3.1.5</a>. Flow ................................................<a href="#page-7">7</a>
68 <a href="#section-3.2">3.2</a>. Measurement Terms ..........................................<a href="#page-7">7</a>
69 <a href="#section-3.2.1">3.2.1</a>. Forwarding Capacity .................................<a href="#page-7">7</a>
70 <a href="#section-3.2.2">3.2.2</a>. Conforming Packet ...................................<a href="#page-8">8</a>
71 <a href="#section-3.2.3">3.2.3</a>. Nonconforming Packet ................................<a href="#page-9">9</a>
72 <a href="#section-3.2.4">3.2.4</a>. Forwarding Delay ....................................<a href="#page-9">9</a>
73 <a href="#section-3.2.5">3.2.5</a>. Jitter .............................................<a href="#page-11">11</a>
74 <a href="#section-3.2.6">3.2.6</a>. Undifferentiated Response ..........................<a href="#page-11">11</a>
75 <a href="#section-3.3">3.3</a>. Sequence Tracking .........................................<a href="#page-12">12</a>
76 <a href="#section-3.3.1">3.3.1</a>. Test Sequence Number ...............................<a href="#page-12">12</a>
77 <a href="#section-3.3.2">3.3.2</a>. Stream .............................................<a href="#page-12">12</a>
78 <a href="#section-3.3.3">3.3.3</a>. In-Sequence Packet .................................<a href="#page-13">13</a>
79 <a href="#section-3.3.4">3.3.4</a>. Out-of-Order Packet ................................<a href="#page-14">14</a>
80 <a href="#section-3.3.5">3.3.5</a>. Duplicate Packet ...................................<a href="#page-14">14</a>
81 <a href="#section-3.4">3.4</a>. Vectors ...................................................<a href="#page-15">15</a>
82 <a href="#section-3.4.1">3.4.1</a>. Intended Vector ....................................<a href="#page-15">15</a>
83 <a href="#section-3.4.2">3.4.2</a>. Offered Vector .....................................<a href="#page-16">16</a>
84 <a href="#section-3.4.3">3.4.3</a>. Expected Vectors ...................................<a href="#page-16">16</a>
85 <a href="#section-3.4.4">3.4.4</a>. Output Vectors .....................................<a href="#page-23">23</a>
86 <a href="#section-4">4</a>. Security Considerations ........................................<a href="#page-30">30</a>
87 <a href="#section-5">5</a>. Acknowledgements ...............................................<a href="#page-30">30</a>
88 <a href="#section-6">6</a>. References .....................................................<a href="#page-31">31</a>
89 <a href="#section-6.1">6.1</a>. Normative References ......................................<a href="#page-31">31</a>
90 <a href="#section-6.2">6.2</a>. Informative References ....................................<a href="#page-31">31</a>
91
92 <span class="h2"><a class="selflink" id="section-1" href="#section-1">1</a>. Introduction</span>
93
94 New terminology is needed because most existing measurements assume
95 the absence of congestion and only a single per-hop behavior. This
96 document introduces several new terms that will allow measurements to
97 be taken during periods of congestion.
98
99 Another key difference from existing terminology is the definition of
100 measurements as observed on egress and ingress of a device/system
101 under test. Again, the existence of congestion requires the addition
102 of egress measurements, as well as of those taken on ingress; without
103 observing traffic leaving a device/system, it is not possible to say
104 whether traffic-control mechanisms effectively dealt with congestion.
105
106
107
108 <span class="grey">Poretsky, et al. Informational [Page 2]</span></pre>
109 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-3" ></span>
110 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
111
112
113 The principal measurements introduced in this document are vectors
114 for rate, delay, and jitter, all of which can be observed with or
115 without congestion of the Device Under Test (DUT)/System Under Test
116 (SUT). This document describes only those terms relevant to
117 measuring behavior of a DUT or SUT at the egress during periods of
118 congestion. End-to-end and service-level measurements are beyond the
119 scope of this document.
120
121 <span class="h2"><a class="selflink" id="section-2" href="#section-2">2</a>. Existing Definitions</span>
122
123 <a href="./rfc1224">RFC 1224</a>, "Techniques for Managing Asynchronously Generated Alerts"
124 [<a href="#ref-St91" title="&quot;Techniques for managing asynchronously generated alerts&quot;">St91</a>], is used for 'Time with fine enough units to distinguish
125 between two events'.
126
127 <a href="./rfc1242">RFC 1242</a>, "Benchmarking Terminology for Network Interconnect
128 Devices", and <a href="./rfc2285">RFC 2285</a>, "Benchmarking Terminology for LAN Switching
129 Devices", should be consulted before attempting to make use of this
130 document.
131
132 <a href="./rfc2474">RFC 2474</a>, "Definition of the Differentiated Services Field (DS Field)
133 in the IPv4 and IPv6 Headers", <a href="#section-2">section 2</a>, contains discussions of a
134 number of terms relevant to network-layer traffic control mechanisms
135 and should also be consulted.
136
137 For the sake of clarity and continuity, this RFC adopts the template
138 for definitions set out in <a href="./rfc1242#section-2">Section&nbsp;2 of RFC 1242</a>. Definitions are
139 indexed and grouped together in sections for ease of reference.
140
141 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
142 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
143 document are to be interpreted as described in <a href="https://www.rfc-editor.org/bcp/bcp14">BCP 14</a>, <a href="./rfc2119">RFC 2119</a>
144 [<a href="#ref-Br97" title="&quot;Key words for use in RFCs to Indicate Requirement Levels&quot;">Br97</a>]. <a href="./rfc2119">RFC 2119</a> defines the use of these key words to help make the
145 intent of standards track documents as clear as possible. While this
146 document uses these keywords, this document is not a standards track
147 document.
148
149 <span class="h3"><a class="selflink" id="section-2.1" href="#section-2.1">2.1</a>. Frequently Used Acronyms</span>
150
151 DA Destination Address
152 DS DiffServ
153 DSCP DiffServ Code Point
154 DUT Device Under Test
155 IP Internet Protocol
156 PHB Per Hop Behavior
157 SA Source Address
158 SUT System Under Test
159
160
161
162
163
164 <span class="grey">Poretsky, et al. Informational [Page 3]</span></pre>
165 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-4" ></span>
166 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
167
168
169 <span class="h2"><a class="selflink" id="section-3" href="#section-3">3</a>. Term Definitions</span>
170
171 <span class="h3"><a class="selflink" id="section-3.1" href="#section-3.1">3.1</a>. Configuration Terms</span>
172
173 <span class="h4"><a class="selflink" id="section-3.1.1" href="#section-3.1.1">3.1.1</a>. Classification</span>
174
175 Definition:
176 Selection of packets according to defined rules.
177
178 Discussion:
179 Classification determines the per-hop behaviors and traffic
180 conditioning functions, such as shaping and dropping, that are to
181 be applied to the packet.
182
183 Classification of packets can be based on the DS field or IP
184 Precedence in the packet header. Classification can be based on
185 other IP header fields, such as IP Source Address (SA),
186 Destination Address (DA), and protocol, or on fields in the packet
187 payload, such as port number. Classification can also be based on
188 ingress interface. It is possible to base classification on
189 Multi-Field (MF) criteria such as IP source and destination
190 addresses, protocol, and port number. For further discussion of
191 packet classification and its network applications, see [<a href="#ref-Bl98" title="&quot;An Architecture for Differentiated Service&quot;">Bl98</a>].
192
193 Measurement units:
194 n/a
195
196 See Also:
197 None
198
199 <span class="h4"><a class="selflink" id="section-3.1.2" href="#section-3.1.2">3.1.2</a>. Codepoint Set</span>
200
201 Definition:
202 The set of all DS Code-points or IP precedence values used during
203 the test duration.
204
205 Discussion:
206 Describes all the code-point markings associated with packets that
207 are input to the DUT/SUT. For each entry in the codepoint set,
208 there are associated vectors describing the rate of traffic,
209 delay, loss, or jitter containing that particular DSCP or IP
210 precedence value.
211
212 The treatment that a packet belonging to a particular code-point
213 gets is subject to the DUT classifying packets to map to the
214 correct PHB. Moreover, the forwarding treatment in general is
215 also dependent on the complete set of offered vectors.
216
217
218
219
220 <span class="grey">Poretsky, et al. Informational [Page 4]</span></pre>
221 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-5" ></span>
222 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
223
224
225 Measurement Units:
226 n/a
227
228 See Also:
229 None
230
231 <span class="h4"><a class="selflink" id="section-3.1.3" href="#section-3.1.3">3.1.3</a>. Forwarding Congestion</span>
232
233 Definition:
234 A condition in which one or more egress interfaces are offered
235 more packets than are forwarded.
236
237 Discussion:
238 This condition is a superset of the overload definition [<a href="#ref-Ma98" title="&quot;Benchmarking Terminology for LAN Switching Devices&quot;">Ma98</a>].
239 Overload [<a href="#ref-Ma98" title="&quot;Benchmarking Terminology for LAN Switching Devices&quot;">Ma98</a>] deals with overloading input and output interfaces
240 beyond the maximum transmission allowed by the medium. Forwarding
241 congestion does not assume ingress interface overload as the only
242 source of overload on output interfaces.
243
244 Another difference between Forwarding Congestion and overload
245 occurs when the SUT comprises multiple elements, in that
246 Forwarding Congestion may occur at multiple points. Consider an
247 SUT comprising multiple edge devices exchanging traffic with a
248 single core device. Depending on traffic patterns, the edge
249 devices may induce Forwarding Congestion on multiple egress
250 interfaces on the core device.
251
252 Throughput [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>] defines the lower boundary of Forwarding
253 Congestion. Throughput is the maximum offered rate with no
254 Forwarding Congestion. At offered rates above throughput, the
255 DUT/SUT is considered to be in a state of Forwarding Congestion.
256
257 Packet Loss, not increased Forwarding Delay, is the external
258 observable metric used to indicate the condition of Forwarding
259 Congestion. Packet Loss is a deterministic indicator of
260 Forwarding Congestion. The condition of increased Forwarding
261 Delay without Packet Loss is an indicator of Forwarding Congestion
262 known as Incipient Congestion. Incipient Congestion is a non-
263 deterministic indicator of Forwarding Congestion [<a href="#ref-Fl93" title="&quot;Random Early Detection gateways for Congestion Avoidance&quot;">Fl93</a>]. As
264 stated in [<a href="#ref-Ec98">Ec98</a>], RED [<a href="#ref-Br98" title="&quot;Recommendations on Queue Management and Congestion Avoidance in the Internet&quot;">Br98</a>] detects incipient congestion before
265 the buffer overflows, but the current Internet environment is
266 limited to packet loss as the mechanism for indicating congestion
267 to the end-nodes. [<a href="#ref-Ra99" title="&quot;The Addition of Explicit Congestion Notification (ECN) to IP&quot;">Ra99</a>] implies that it is impractical to build
268 a black-box test to observe Incipient Congestion. [<a href="#ref-Ra99" title="&quot;The Addition of Explicit Congestion Notification (ECN) to IP&quot;">Ra99</a>] instead
269 introduces Explicit Congestion Notification (ECN) as a
270 deterministic Black-Box method for observing Incipient Congestion.
271 [<a href="#ref-Ra99" title="&quot;The Addition of Explicit Congestion Notification (ECN) to IP&quot;">Ra99</a>] is an Experimental RFC with limited deployment, so ECN is
272 not used for this particular methodology. For the purpose of
273
274
275
276 <span class="grey">Poretsky, et al. Informational [Page 5]</span></pre>
277 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-6" ></span>
278 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
279
280
281 "black-box" testing a DUT/SUT, this methodology uses Packet Loss
282 as the indicator of Forwarding Congestion.
283
284 Ingress observations alone are not sufficient to cover all cases
285 in which Forwarding Congestion may occur. A device with an
286 infinite amount of memory could buffer an infinite number of
287 packets and eventually forward all of them. However, these
288 packets may or may not be forwarded during the test duration.
289 Congestion Collapse [<a href="#ref-Na84" title="&quot;Congestion control in IP/TCP internetworks&quot;">Na84</a>] is defined as the state in which
290 buffers are full and all arriving packets MUST be dropped across
291 the network. Even though ingress interfaces accept all packets
292 without loss, Forwarding Congestion is present in this
293 hypothetical device.
294
295 The definition presented here explicitly defines Forwarding
296 Congestion as an event observable on egress interfaces.
297 Regardless of internal architecture, any device exhibiting Packet
298 Loss on one or more egress interfaces is experiencing Forwarding
299 Congestion.
300
301 Measurement units:
302 None
303
304 See Also:
305 Gateway Congestion Control Survey [<a href="#ref-Ma91" title="&quot;Gateway Congestion Control Survey&quot;">Ma91</a>]
306
307 <span class="h4"><a class="selflink" id="section-3.1.4" href="#section-3.1.4">3.1.4</a>. Congestion Management</span>
308
309 Definition:
310 An implementation of one or more per-hop behaviors to avoid or
311 minimize the condition of congestion.
312
313 Discussion:
314 Congestion management may seek either to control congestion or
315 avoid it altogether through Classification.
316
317 Congestion avoidance mechanisms seek to prevent congestion before
318 it actually occurs.
319
320 Congestion control mechanisms give one or more flows (with a
321 discrete IP Precedence or DSCP value) preferential treatment over
322 other classes during periods of congestion.
323
324 Measurement units:
325 n/a
326
327 See Also:
328 Classification
329
330
331
332 <span class="grey">Poretsky, et al. Informational [Page 6]</span></pre>
333 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-7" ></span>
334 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
335
336
337 <span class="h4"><a class="selflink" id="section-3.1.5" href="#section-3.1.5">3.1.5</a>. Flow</span>
338
339 Definition:
340 A flow is one or more packets sharing a common intended pair of
341 ingress and egress interfaces.
342
343 Discussion:
344 Packets are grouped by the ingress and egress interfaces they use
345 on a given DUT/SUT.
346
347 A flow can contain multiple source IP addresses and/or destination
348 IP addresses. All packets in a flow MUST enter on the same
349 ingress interface and exit on the same egress interface and have
350 some common network layer content.
351
352 Microflows [<a href="#ref-Ni98" title="&quot;Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers&quot;">Ni98</a>] are a subset of flows. As defined in [<a href="#ref-Ni98" title="&quot;Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers&quot;">Ni98</a>],
353 microflows require application-to-application measurement. In
354 contrast, flows use lower-layer classification criteria. Since
355 this document focuses on network-layer classification criteria, it
356 concentrates here on the use of network-layer identifiers in
357 describing a flow. Flow identifiers also may reside at the data-
358 link, transport, or application layers of the OSI model. However,
359 identifiers other than those at the network layer are out of scope
360 for this document.
361
362 A flow may contain a single code point/IP precedence value or may
363 contain multiple values destined for a single egress interface.
364 This is determined by the test methodology.
365
366 Measurement units:
367 n/a
368
369 See Also:
370 Microflow [<a href="#ref-Ni98" title="&quot;Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers&quot;">Ni98</a>]
371 Streams
372
373 <span class="h3"><a class="selflink" id="section-3.2" href="#section-3.2">3.2</a>. Measurement Terms</span>
374
375 <span class="h4"><a class="selflink" id="section-3.2.1" href="#section-3.2.1">3.2.1</a>. Forwarding Capacity</span>
376
377 Definition:
378 The number of packets per second that a device can be observed to
379 transmit successfully to the correct egress interface in response
380 to a specified offered load while the device drops none of the
381 offered packets.
382
383
384
385
386
387
388 <span class="grey">Poretsky, et al. Informational [Page 7]</span></pre>
389 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-8" ></span>
390 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
391
392
393 Discussion:
394 Forwarding Capacity measures the packet rate at the egress
395 interface(s) of the DUT/SUT. In contrast, throughput (as defined
396 in <a href="./rfc1242">RFC 1242</a>) measures the packet rate at the ingress interface(s)
397 of the DUT/SUT.
398
399 Ingress-based measurements do not account for queuing of the
400 DUT/SUT. Throughput rates can be higher than the Forwarding
401 Capacity because of queueing. The difference is dependent upon
402 test duration, packet rate, and queue size. Forwarding Capacity,
403 as an egress measurement, does take queuing into account.
404
405 Understanding Forwarding Capacity is a necessary precursor to any
406 measurement involving Traffic Control Mechanisms. The
407 accompanying methodology document MUST take into consideration
408 Forwarding Capacity when determining the expected forwarding
409 vectors. When the sum of the expected forwarding vectors on an
410 interface exceeds the Forwarding Capacity, the Forwarding Capacity
411 will govern the forwarding rate.
412
413 This measurement differs from forwarding rate at maximum offered
414 load (FRMOL) [<a href="#ref-Ma98" title="&quot;Benchmarking Terminology for LAN Switching Devices&quot;">Ma98</a>] in that the Forwarding Capacity requires zero
415 loss.
416
417 Measurement units:
418 N-octet packets per second
419
420 See Also:
421 Throughput [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>]
422 Forwarding Rate at Maximum Offered Load [<a href="#ref-Ma98" title="&quot;Benchmarking Terminology for LAN Switching Devices&quot;">Ma98</a>]
423
424 <span class="h4"><a class="selflink" id="section-3.2.2" href="#section-3.2.2">3.2.2</a>. Conforming Packet</span>
425
426 Definition:
427 Packets that lie within specific rate, delay, or jitter bounds.
428
429 Discussion:
430 A DUT/SUT may be configured to allow a given traffic class to
431 consume a given amount of bandwidth, or to fall within predefined
432 delay or jitter boundaries. All packets that lie within specified
433 bounds are then said to be conforming, whereas those outside the
434 bounds are nonconforming.
435
436 Measurement units:
437 n/a
438
439
440
441
442
443
444 <span class="grey">Poretsky, et al. Informational [Page 8]</span></pre>
445 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-9" ></span>
446 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
447
448
449 See Also:
450 Expected Vector
451 Forwarding Vector
452 Offered Vector
453 Nonconforming
454
455 <span class="h4"><a class="selflink" id="section-3.2.3" href="#section-3.2.3">3.2.3</a>. Nonconforming Packet</span>
456
457 Definition:
458 Packets that do not lie within specific rate, delay, or jitter
459 bounds.
460
461 Discussion:
462 A DUT/SUT may be configured to allow a given traffic class to
463 consume a given amount of bandwidth, or to fall within predefined
464 delay or jitter boundaries. All packets that do not lie within
465 these bounds are then said to be nonconforming.
466
467 Measurement units:
468 n/a
469
470 See Also:
471 Expected Vector
472 Forwarding Vector
473 Offered Vector
474 Conforming
475
476 <span class="h4"><a class="selflink" id="section-3.2.4" href="#section-3.2.4">3.2.4</a>. Forwarding Delay</span>
477
478 Definition:
479 The time interval starting when the last bit of the input IP
480 packet is offered to the input port of the DUT/SUT and ending when
481 the last bit of the output IP packet is received from the output
482 port of the DUT/SUT.
483
484 Discussion:
485 The delay time interval MUST be externally observed. The delay
486 measurement MUST NOT include delays added by test bed components
487 other than the DUT/SUT, such as propagation time introduced by
488 cabling or non-zero delay added by the test instrument.
489 Forwarding Delay differs from latency [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>] and one-way delay
490 [<a href="#ref-Al99" title="&quot;A One-way Delay Metric for IPPM&quot;">Al99</a>] in several key regards:
491
492 1. Latency [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>] assumes knowledge of whether the DUT/SUT uses
493 "store and forward" or "bit forwarding" technology. Forwarding
494 Delay is the same metric, measured the same way, regardless of
495 the architecture of the DUT/SUT.
496
497
498
499
500 <span class="grey">Poretsky, et al. Informational [Page 9]</span></pre>
501 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-10" ></span>
502 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
503
504
505 2. Forwarding Delay is a last-in, last-out (LILO) measurement,
506 unlike the last-in, first-out method [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>] or the first-in,
507 last-out method [<a href="#ref-Al99" title="&quot;A One-way Delay Metric for IPPM&quot;">Al99</a>].
508
509 The LILO method most closely simulates the way a network-layer
510 device actually processes an IP datagram. IP datagrams are not
511 passed up and down the stack unless they are complete, and
512 processing begins only once the last bit of the IP datagram has
513 been received.
514
515 Further, the LILO method has an additive property, where the
516 sum of the parts MUST equal the whole. This is a key
517 difference from [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>] and [<a href="#ref-Al99" title="&quot;A One-way Delay Metric for IPPM&quot;">Al99</a>]. For example, the delay
518 added by two DUTs MUST equal the sum of the delay of the DUTs.
519 This may or may not be the case with [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>] and [<a href="#ref-Al99" title="&quot;A One-way Delay Metric for IPPM&quot;">Al99</a>].
520
521 3. Forwarding Delay measures the IP datagram only, unlike [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>],
522 which also includes link-layer overhead.
523
524 A metric focused exclusively on the Internet protocol relieves
525 the tester from specifying the start/end for every link-layer
526 protocol that IP runs on. This avoids the need to determine
527 whether the start/stop delimiters are included. It also allows
528 the use of heterogeneous link-layer protocols in a test.
529
530 4. Forwarding Delay can be measured at any offered load, whereas
531 the latency methodology [<a href="#ref-Br99" title="&quot;Benchmarking Methodology for Network Interconnect Devices&quot;">Br99</a>] recommends measurement at, and
532 only at, the throughput level. Comparing the Forwarding Delay
533 below the throughput to Forwarding Delay above the Forwarding
534 Capacity will give insight to the traffic control mechanisms.
535
536 For example, non-congested delay may be measured with an
537 offered load that does not exceed the Forwarding Capacity,
538 while congested delay may involve an offered load that exceeds
539 the Forwarding Capacity.
540
541 Note: Forwarding Delay SHOULD NOT be used as an absolute
542 indicator of DUT/SUT Forwarding Congestion. While Forwarding
543 Delay may rise when offered load nears or exceeds the
544 Forwarding Capacity, there is no universal point at which
545 Forwarding Delay can be said to indicate the presence or
546 absence of Forwarding Congestion.
547
548 Measurement units:
549 milliseconds
550
551
552
553
554
555
556 <span class="grey">Poretsky, et al. Informational [Page 10]</span></pre>
557 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-11" ></span>
558 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
559
560
561 See Also:
562 Latency [<a href="#ref-Br91" title="&quot;Benchmarking terminology for network interconnection devices&quot;">Br91</a>]
563 Latency [<a href="#ref-Al99" title="&quot;A One-way Delay Metric for IPPM&quot;">Al99</a>]
564 One-way Delay [<a href="#ref-Br99" title="&quot;Benchmarking Methodology for Network Interconnect Devices&quot;">Br99</a>]
565
566 <span class="h4"><a class="selflink" id="section-3.2.5" href="#section-3.2.5">3.2.5</a>. Jitter</span>
567
568 Definition:
569 The absolute value of the difference between the Forwarding Delay
570 of two consecutive received packets belonging to the same stream.
571
572 Discussion:
573 The Forwarding Delay fluctuation between two consecutive received
574 packets in a stream is reported as the jitter. Jitter can be
575 expressed as |D(i) - D(i-1)|, where D equals the Forwarding Delay
576 and i is the order the packets were received.
577
578 Under loss, jitter can be measured between non-consecutive test
579 sequence numbers. When IP Traffic Control Mechanisms are dropping
580 packets, fluctuating Forwarding Delay may be observed. Jitter
581 MUST be able to benchmark the delay variation independently of
582 packet loss.
583
584 Jitter is related to the IPDV [<a href="#ref-De02" title="&quot;IP Packet Delay Variation Metric for IP Performance Metrics (IPPM)&quot;">De02</a>] (IP Delay Variation) by
585 taking the absolute value of the ipdv. The two metrics will
586 produce different mean values. Mean Jitter will produce a
587 positive value, where the mean ipdv is typically zero. Also, IPDV
588 is undefined when one packet from a pair is lost.
589
590 Measurement units:
591 milliseconds
592
593 See Also:
594 Forwarding Delay
595 Jitter variation [<a href="#ref-Ja99" title="&quot;An Expedited Forwarding PHB (Per-Hop Behavior)&quot;">Ja99</a>]
596 ipdv [<a href="#ref-De02" title="&quot;IP Packet Delay Variation Metric for IP Performance Metrics (IPPM)&quot;">De02</a>]
597 interarrival jitter [<a href="#ref-Sc96" title="&quot;RTP: A Transport Protocol for Real-Time Applications&quot;">Sc96</a>]
598
599 <span class="h4"><a class="selflink" id="section-3.2.6" href="#section-3.2.6">3.2.6</a>. Undifferentiated Response</span>
600
601 Definition:
602 The vector(s) obtained when mechanisms used to support diff-serv
603 or IP precedence are disabled.
604
605 Discussion:
606 Enabling diff-serv or IP precedence mechanisms may impose
607 additional processing overhead for packets. This overhead may
608 degrade performance even when traffic belonging to only one class,
609
610
611
612 <span class="grey">Poretsky, et al. Informational [Page 11]</span></pre>
613 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-12" ></span>
614 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
615
616
617 the best-effort class, is offered to the device. Measurements
618 with "undifferentiated response" SHOULD be made to establish a
619 baseline.
620
621 The vector(s) obtained with DSCP or IP precedence enabled can be
622 compared to the undifferentiated response to determine the effect
623 of differentiating traffic.
624
625 Measurement units:
626 n/a
627
628 <span class="h3"><a class="selflink" id="section-3.3" href="#section-3.3">3.3</a>. Sequence Tracking</span>
629
630 <span class="h4"><a class="selflink" id="section-3.3.1" href="#section-3.3.1">3.3.1</a>. Test Sequence Number</span>
631
632 Definition:
633 A field in the IP payload portion of the packet that is used to
634 verify the order of the packets on the egress of the DUT/SUT.
635
636 Discussion:
637 The traffic generator sets the test sequence number value. Upon
638 receipt of the packet, the traffic receiver checks the value.
639 The traffic generator changes the value on each packet transmitted
640 based on an algorithm agreed to by the traffic receiver.
641
642 The traffic receiver keeps track of the sequence numbers on a
643 per-stream basis. In addition to the number of received packets,
644 the traffic receiver may also report the number of in-sequence
645 packets, the number of out-of-sequence packets, the number of
646 duplicate packets, and the number of reordered packets. The
647 RECOMMENDED algorithm to change the sequence number on sequential
648 packets is an incrementing value.
649
650 Measurement units:
651 n/a
652
653 See Also:
654 Stream
655
656 <span class="h4"><a class="selflink" id="section-3.3.2" href="#section-3.3.2">3.3.2</a>. Stream</span>
657
658 Definition:
659 A group of packets tracked as a single entity by the traffic
660 receiver. A stream MUST share common content, such as type (IP,
661 UDP), IP SA/DA, packet size, or payload.
662
663
664
665
666
667
668 <span class="grey">Poretsky, et al. Informational [Page 12]</span></pre>
669 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-13" ></span>
670 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
671
672
673 Discussion:
674 Streams are tracked by test sequence number or "unique signature
675 field" [<a href="#ref-Ma00" title="&quot;Benchmarking Methodology for LAN Switching Devices&quot;">Ma00</a>]. Streams define how individual packet statistics
676 are grouped together to form an intelligible summary.
677
678 Common stream groupings would be by egress interface, destination
679 address, source address, DSCP, or IP precedence. A stream using
680 test sequence numbers can track the ordering of packets as they
681 traverse the DUT/SUT.
682
683 Streams are not restricted to a pair of source and destination
684 interfaces as long as all packets are tracked as a single entity.
685 A multicast stream can be forwarded to multiple destination
686 interfaces.
687
688 Measurement units:
689 n/a
690
691 See Also:
692 Flow
693 Microflow [<a href="#ref-Ni98" title="&quot;Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers&quot;">Ni98</a>]
694 Test sequence number
695
696 <span class="h4"><a class="selflink" id="section-3.3.3" href="#section-3.3.3">3.3.3</a>. In-Sequence Packet</span>
697
698 Definition:
699 A received packet with the expected Test Sequence number.
700
701 Discussion:
702 In-sequence is done on a stream level. As packets are received on
703 a stream, each packet's Test Sequence number is compared with the
704 previous packet. Only packets that match the expected Test
705 Sequence number are considered in-sequence.
706
707 Packets that do not match the expected Test Sequence number are
708 counted as "not in-sequence" or out-of-sequence. Every packet
709 that is received is either in-sequence or out-of-sequence.
710 Subtracting the in-sequence from the received packets (for that
711 stream), the tester can derive the out-of-sequence count.
712
713 Two types of events will prevent the in-sequence from
714 incrementing: packet loss and reordered packets.
715
716 Measurement units:
717 Packet count
718
719
720
721
722
723
724 <span class="grey">Poretsky, et al. Informational [Page 13]</span></pre>
725 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-14" ></span>
726 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
727
728
729 See Also:
730 Stream
731 Test Sequence number
732
733 <span class="h4"><a class="selflink" id="section-3.3.4" href="#section-3.3.4">3.3.4</a>. Out-of-Order Packet</span>
734
735 Definition:
736 A received packet with a sequence number less than the sequence
737 number of any previously arriving packet.
738
739 Discussion:
740 As a stream of packets enters a DUT/SUT, they include a Stream
741 Test Sequence number indicating the order the packets were sent to
742 the DUT/SUT. On exiting the DUT/SUT, these packets may arrive in
743 a different order. Each packet that was reordered is counted as
744 an Out-of-Order Packet.
745
746 Certain streaming protocols (such as TCP) require the packets to
747 be in a certain order. Packets outside this are dropped by the
748 streaming protocols even though they were properly received by the
749 IP layer. The type of reordering tolerated by a streaming
750 protocol varies from protocol to protocol, and also by
751 implementation.
752
753 Packet loss does not affect the Out-of-Order Packet count. The
754 Out-of-Order Packet count is impacted only by packets that were
755 not received in the order that they were transmitted.
756
757 Measurement units:
758 packets
759
760 See Also:
761 Stream
762 Test Sequence number
763 Packet Reordering Metric for IPPM [<a href="#ref-Mo03" title="&quot;Packet Reordering Metric for IPPM&quot;">Mo03</a>]
764
765 <span class="h4"><a class="selflink" id="section-3.3.5" href="#section-3.3.5">3.3.5</a>. Duplicate Packet</span>
766
767 Definition:
768 A received packet with a Test Sequence number matching a
769 previously received packet.
770
771 Discussion:
772 A Duplicate Packet is a packet that the DUT/SUT has successfully
773 transmitted out an egress interface more than once. The egress
774 interface has previously forwarded this packet.
775
776
777
778
779
780 <span class="grey">Poretsky, et al. Informational [Page 14]</span></pre>
781 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-15" ></span>
782 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
783
784
785 A Duplicate Packet SHOULD be a bit-for-bit copy of an already
786 transmitted packet (including Test Sequence number). If the
787 Duplicate Packet traversed different paths through the DUT/SUT,
788 some fields (such as TTL or checksum) may have changed.
789
790 A multicast packet is not a Duplicate Packet by definition. For a
791 given IP multicast group, a DUT/SUT SHOULD forward a packet once
792 on a given egress interface provided the path to one or more
793 multicast receivers is through that interface. Several egress
794 interfaces will transmit the same packet, but only once per
795 interface.
796
797 To detect a Duplicate Packet, each packet offered to the DUT/SUT
798 MUST contain a unique packet-by-packet identifier.
799
800 Measurement units:
801 Packet count
802
803 See Also:
804 Stream
805 Test Sequence number
806
807 <span class="h3"><a class="selflink" id="section-3.4" href="#section-3.4">3.4</a>. Vectors</span>
808
809 A vector is a group of packets all matching a specific
810 classification criteria, such as DSCP. Vectors are
811 identified by the classification criteria and benchmarking
812 metrics, such as a Forwarding Capacity, Forwarding Delay,
813 or Jitter.
814
815 <span class="h4"><a class="selflink" id="section-3.4.1" href="#section-3.4.1">3.4.1</a>. Intended Vector</span>
816
817 Definition:
818 A description of the configuration on an external source
819 for the attempted rate of a stream transmitted to a DUT/SUT
820 matching specific classification rules.
821
822 Discussion:
823 The Intended Vector of a stream influences the benchmark
824 measurements. The Intended Vector is described by the
825 classification criteria and attempted rate.
826
827 Measurement Units:
828 N-bytes packets per second
829
830
831
832
833
834
835
836 <span class="grey">Poretsky, et al. Informational [Page 15]</span></pre>
837 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-16" ></span>
838 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
839
840
841 See Also:
842 Stream
843 Offered Vector
844 Forwarding Vector
845
846 <span class="h4"><a class="selflink" id="section-3.4.2" href="#section-3.4.2">3.4.2</a>. Offered Vector</span>
847
848 Definition:
849 A description for the attempted rate of a stream offered to
850 a DUT/SUT matching specific classification rules.
851
852 Discussion:
853 The Offered Vector of a stream influences the benchmark
854 measurements. The Offered Vector is described by the
855 classification criteria and offered rate.
856
857 Measurement Units:
858 N-bytes packets per second
859
860 See Also:
861 Stream
862 Intended Vector
863 Forwarding Vector
864
865 <span class="h4"><a class="selflink" id="section-3.4.3" href="#section-3.4.3">3.4.3</a>. Expected Vectors</span>
866
867 <span class="h5"><a class="selflink" id="section-3.4.3.1" href="#section-3.4.3.1">3.4.3.1</a>. Expected Forwarding Vector</span>
868
869 Definition:
870 A description of the expected output rate of packets matching a
871 specific classification, such as DSCP.
872
873 Discussion:
874 The value of the Expected Forwarding Vector is dependent on the
875 set of offered vectors and Classification configuration on the
876 DUT/SUT. The DUT is configured in a certain way so that
877 classification occurs when a traffic mix consisting of multiple
878 streams is applied.
879
880 This term captures the expected forwarding behavior from the DUT
881 receiving multiple Offered Vectors. The actual algorithm or
882 mechanism the DUT uses to achieve service differentiation is
883 implementation specific and is not important when describing the
884 Expected Forwarding Vector.
885
886 Measurement units:
887 N-octet packets per second
888
889
890
891
892 <span class="grey">Poretsky, et al. Informational [Page 16]</span></pre>
893 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-17" ></span>
894 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
895
896
897 See Also:
898 Classification
899 Stream
900 Intended Vector
901 Offered Vector
902
903 <span class="h5"><a class="selflink" id="section-3.4.3.2" href="#section-3.4.3.2">3.4.3.2</a>. Expected Loss Vector</span>
904
905 Definition:
906 A description of the percentage of packets having a specific
907 classification that should not be forwarded.
908
909 Discussion:
910 The value of the Expected Loss Vector is dependent on the set of
911 offered vectors and Classification configuration on the DUT/SUT.
912 The DUT is configured in a certain way so that classification
913 occurs when a traffic mix consisting of multiple streams is
914 applied.
915
916 This term captures the expected forwarding behavior from the DUT
917 receiving multiple Offered Vectors. The actual algorithm or
918 mechanism the DUT uses to achieve service differentiation is
919 implementation specific and is not important when describing the
920 Expected Loss Vector.
921
922 Measurement Units:
923 Percentage of intended packets expected to be dropped.
924
925 See Also:
926 Classification
927 Stream
928 Intended Vector
929 Offered Vector
930 One-way Packet Loss Metric [<a href="#ref-Ka99" title="&quot;A One-way Packet Loss Metric for IPPM&quot;">Ka99</a>]
931
932 <span class="h5"><a class="selflink" id="section-3.4.3.3" href="#section-3.4.3.3">3.4.3.3</a>. Expected Sequence Vector</span>
933
934 Definition:
935 A description of the expected in-sequence packets matching a
936 specific classification, such as DSCP.
937
938 Discussion:
939 The value of the Expected Sequence Vector is dependent on the set
940 of offered vectors and Classification configuration on the
941 DUT/SUT. The DUT is configured in a certain way so that
942 classification occurs when a traffic mix consisting of multiple
943 streams is applied.
944
945
946
947
948 <span class="grey">Poretsky, et al. Informational [Page 17]</span></pre>
949 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-18" ></span>
950 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
951
952
953 This term captures the expected forwarding behavior from the DUT
954 receiving multiple Offered Vectors. The actual algorithm or
955 mechanism the DUT uses to achieve service differentiation is
956 implementation specific and is not important when describing the
957 Expected Sequence Vector.
958
959 Measurement Units:
960 N-octet packets per second
961
962 See Also:
963 Classification
964 Stream
965 In-Sequence Packet
966 Intended Vector
967 Offered Vector
968
969 <span class="h5"><a class="selflink" id="section-3.4.3.4" href="#section-3.4.3.4">3.4.3.4</a>. Expected Delay Vector</span>
970
971 Definition:
972 A description of the expected instantaneous Forwarding Delay for
973 packets matching a specific classification, such as DSCP.
974
975 Discussion:
976 The value of the Expected Delay Vector is dependent on the set of
977 offered vectors and Classification configuration on the DUT/SUT.
978 The DUT is configured in a certain way so that classification
979 occurs when a traffic mix consisting of multiple streams is
980 applied.
981
982 This term captures the expected forwarding behavior from the DUT
983 receiving multiple Offered Vectors. The actual algorithm or
984 mechanism the DUT uses to achieve service differentiation is
985 implementation specific and is not important when describing the
986 Expected Delay Vector.
987
988 Measurement units:
989 milliseconds
990
991 See Also:
992 Classification
993 Stream
994 Forwarding Delay
995 Intended Vector
996 Offered Vector
997
998
999
1000
1001
1002
1003
1004 <span class="grey">Poretsky, et al. Informational [Page 18]</span></pre>
1005 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-19" ></span>
1006 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1007
1008
1009 <span class="h5"><a class="selflink" id="section-3.4.3.5" href="#section-3.4.3.5">3.4.3.5</a>. Expected Average Delay Vector</span>
1010
1011 Definition:
1012 A description of the expected average Forwarding Delay for packets
1013 matching a specific classification, such as DSCP.
1014
1015 Discussion:
1016 The value of the Expected Average Delay Vector is dependent on the
1017 set of offered vectors and Classification configuration on the
1018 DUT/SUT. The DUT is configured in a certain way so that
1019 classification occurs when a traffic mix consisting of multiple
1020 streams is applied.
1021
1022 This term captures the expected forwarding behavior from the DUT
1023 receiving multiple Offered Vectors. The actual algorithm or
1024 mechanism the DUT uses to achieve service differentiation is
1025 implementation specific and is not important when describing the
1026 Expected Average Delay Vector.
1027
1028 Measurement units:
1029 milliseconds
1030
1031 See Also:
1032 Classification
1033 Stream
1034 Forwarding Delay
1035 Intended Vector
1036 Offered Vector
1037 Expected Delay Vector
1038
1039 <span class="h5"><a class="selflink" id="section-3.4.3.6" href="#section-3.4.3.6">3.4.3.6</a>. Expected Maximum Delay Vector</span>
1040
1041 Definition:
1042 A description of the expected maximum Forwarding Delay for packets
1043 matching a specific classification, such as DSCP.
1044
1045 Discussion:
1046 The value of the Expected Maximum Delay Vector is dependent on the
1047 set of offered vectors and Classification configuration on the
1048 DUT/SUT. The DUT is configured in a certain way so that
1049 classification occurs when a traffic mix consisting of multiple
1050 streams is applied.
1051
1052 This term captures the expected forwarding behavior from the DUT
1053 receiving multiple Offered Vectors. The actual algorithm or
1054 mechanism the DUT uses to achieve service differentiation is
1055 implementation specific and is not important when describing the
1056 Expected Maximum Delay Vector.
1057
1058
1059
1060 <span class="grey">Poretsky, et al. Informational [Page 19]</span></pre>
1061 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-20" ></span>
1062 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1063
1064
1065 Measurement units:
1066 milliseconds
1067
1068 See Also:
1069 Classification
1070 Stream
1071 Forwarding Delay
1072 Intended Vector
1073 Offered Vector
1074 Expected Delay Vector
1075
1076 <span class="h5"><a class="selflink" id="section-3.4.3.7" href="#section-3.4.3.7">3.4.3.7</a>. Expected Minimum Delay Vector</span>
1077
1078 Definition:
1079 A description of the expected minimum Forwarding Delay for packets
1080 matching a specific classification, such as DSCP.
1081
1082 Discussion:
1083 The value of the Expected Minimum Delay Vector is dependent on the
1084 set of offered vectors and Classification configuration on the
1085 DUT/SUT. The DUT is configured in a certain way so that
1086 classification occurs when a traffic mix consisting of multiple
1087 streams is applied.
1088
1089 This term captures the expected forwarding behavior from the DUT
1090 receiving multiple Offered Vectors. The actual algorithm or
1091 mechanism the DUT uses to achieve service differentiation is
1092 implementation specific and is not important when describing the
1093 Expected Minimum Delay Vector.
1094
1095 Measurement units:
1096 milliseconds
1097
1098 See Also:
1099 Classification
1100 Stream
1101 Forwarding Delay
1102 Intended Vector
1103 Offered Vector
1104 Expected Delay Vector
1105
1106 <span class="h5"><a class="selflink" id="section-3.4.3.8" href="#section-3.4.3.8">3.4.3.8</a>. Expected Instantaneous Jitter Vector</span>
1107
1108 Definition:
1109 A description of the expected Instantaneous Jitter between two
1110 consecutive packets arrival times matching a specific
1111 classification, such as DSCP.
1112
1113
1114
1115
1116 <span class="grey">Poretsky, et al. Informational [Page 20]</span></pre>
1117 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-21" ></span>
1118 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1119
1120
1121 Discussion:
1122 Instantaneous Jitter is the absolute value of the difference
1123 between the Forwarding Delay measurement of two packets belonging
1124 to the same stream.
1125
1126 The Forwarding Delay fluctuation between two consecutive packets
1127 in a stream is reported as the "Instantaneous Jitter".
1128 Instantaneous Jitter can be expressed as |D(i) - D(i-1)|, where D
1129 equals the Forwarding Delay and i is the test sequence number.
1130 Packets lost are not counted in the measurement.
1131
1132 The Forwarding Vector may contain several Jitter Vectors. For n
1133 packets received in a Forwarding Vector, there is a total of (n-1)
1134 Instantaneous Jitter Vectors.
1135
1136 Measurement units:
1137 milliseconds
1138
1139 See Also:
1140 Classification
1141 Stream
1142 Jitter
1143 Intended Vector
1144 Offered Vector
1145
1146 <span class="h5"><a class="selflink" id="section-3.4.3.9" href="#section-3.4.3.9">3.4.3.9</a>. Expected Average Jitter Vector</span>
1147
1148 Definition:
1149 A description of the expected average jitter for packets arriving
1150 in a stream matching a specific classification, such as DSCP.
1151
1152 Discussion:
1153 Average Jitter Vector is the average of all the Instantaneous
1154 Jitter Vectors measured during the test duration for the same
1155 stream.
1156
1157 The value of the Expected Average Jitter Vector is dependent on
1158 the set of offered vectors and Classification configuration on the
1159 DUT/SUT. The DUT is configured in a certain way so that
1160 classification occurs when a traffic mix consisting of multiple
1161 streams is applied.
1162
1163 This term captures the expected forwarding behavior from the DUT
1164 receiving multiple Offered Vectors. The actual algorithm or
1165 mechanism the DUT uses to achieve service differentiation is
1166 implementation specific and is not important when describing the
1167 Expected Average Jitter Vector.
1168
1169
1170
1171
1172 <span class="grey">Poretsky, et al. Informational [Page 21]</span></pre>
1173 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-22" ></span>
1174 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1175
1176
1177 Measurement units:
1178 milliseconds
1179
1180 See Also:
1181 Classification
1182 Stream
1183 Jitter
1184 Intended Vector
1185 Offered Vector
1186 Expected Instantaneous Jitter Vector
1187
1188 <span class="h5"><a class="selflink" id="section-3.4.3.10" href="#section-3.4.3.10">3.4.3.10</a>. Expected Peak-to-peak Jitter Vector</span>
1189
1190 Definition:
1191 A description of the expected maximum variation in the Forwarding
1192 Delay of packet arrival times for packets arriving in a stream
1193 matching a specific classification, such as DSCP.
1194
1195 Discussion:
1196 Peak-to-peak Jitter Vector is the maximum Forwarding Delay minus
1197 the minimum Forwarding Delay of the packets (in a vector)
1198 forwarded by the DUT/SUT.
1199
1200 Peak-to-peak Jitter is not derived from the Instantaneous Jitter
1201 Vector. Peak-to-peak Jitter is based upon all the packets during
1202 the test duration, not just two consecutive packets.
1203
1204 The value of the Expected Peak-to-peak Jitter Vector is dependent
1205 on the set of offered vectors and Classification configuration on
1206 the DUT/SUT. The DUT is configured in a certain way so that
1207 classification occurs when a traffic mix consisting of multiple
1208 streams is applied.
1209
1210 This term captures the expected forwarding behavior from the DUT
1211 receiving multiple Offered Vectors. The actual algorithm or
1212 mechanism the DUT uses to achieve service differentiation is
1213 implementation specific and is not important when describing the
1214 Expected Peak-to-peak Jitter Vector.
1215
1216 Measurement units:
1217 milliseconds
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228 <span class="grey">Poretsky, et al. Informational [Page 22]</span></pre>
1229 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-23" ></span>
1230 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1231
1232
1233 See Also:
1234 Classification
1235 Stream
1236 Jitter
1237 Intended Vector
1238 Offered Vector
1239 Expected Instantaneous Jitter Vector
1240 Expected Average Jitter Vector
1241
1242 <span class="h4"><a class="selflink" id="section-3.4.4" href="#section-3.4.4">3.4.4</a>. Output Vectors</span>
1243
1244 <span class="h5"><a class="selflink" id="section-3.4.4.1" href="#section-3.4.4.1">3.4.4.1</a>. Forwarding Vector</span>
1245
1246 Definition:
1247 The number of packets per second for a stream matching a specific
1248 classification, such as DSCP, that a DUT/SUT is measured to
1249 forward to the correct destination interface successfully in
1250 response to an offered vector.
1251
1252 Discussion:
1253 Forwarding Vector is expressed as a combination of values: the
1254 classification rules AND the measured packets per second for the
1255 stream matching the classification rules. Forwarding Vector is a
1256 per-hop measurement. The DUT/SUT MAY remark the specific DSCP (or
1257 IP precedence) value for a multi-hop measurement. The stream
1258 remains the same.
1259
1260 Measurement units:
1261 N-octet packets per second
1262
1263 See Also:
1264 Classification
1265 Stream
1266 Forwarding Capacity
1267 Intended Vector
1268 Offered Vector
1269 Expected Vector
1270
1271 <span class="h5"><a class="selflink" id="section-3.4.4.2" href="#section-3.4.4.2">3.4.4.2</a>. Loss Vector</span>
1272
1273 Definition:
1274 The percentage of packets per second for a stream matching a
1275 specific classification, such as DSCP, that a DUT/SUT is measured
1276 not to transmit to the correct destination interface in response
1277 to an offered vector.
1278
1279
1280
1281
1282
1283
1284 <span class="grey">Poretsky, et al. Informational [Page 23]</span></pre>
1285 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-24" ></span>
1286 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1287
1288
1289 Discussion:
1290 Loss Vector is expressed as a combination of values: the
1291 classification rules AND the measured percentage value of packet
1292 loss. Loss Vector is a per-hop measurement. The DUT/SUT MAY
1293 remark the specific DSCP or IP precedence value for a multi-hop
1294 measurement. The stream remains the same.
1295
1296 Measurement Units:
1297 Percentage of packets
1298
1299 See Also:
1300 Classification
1301 Stream
1302 Intended Vector
1303 Offered Vector
1304 Expected Vector
1305 One-way Packet Loss Metric [<a href="#ref-Ka99" title="&quot;A One-way Packet Loss Metric for IPPM&quot;">Ka99</a>]
1306
1307 <span class="h5"><a class="selflink" id="section-3.4.4.3" href="#section-3.4.4.3">3.4.4.3</a>. Sequence Vector</span>
1308
1309 Definition:
1310 The number of packets per second for all packets in a stream
1311 matching a specific classification, such as DSCP, that a DUT/SUT
1312 is measured to transmit in sequence to the correct destination
1313 interface in response to an offered vector.
1314
1315 Discussion:
1316 Sequence Vector is expressed as a combination of values: the
1317 classification rules AND the number of packets per second that are
1318 in-sequence.
1319
1320 Sequence Vector is a per-hop measurement. The DUT/SUT MAY remark
1321 the specific DSCP or IP precedence value for a multi-hop
1322 measurement. The stream remains the same.
1323
1324 Measurement Units:
1325 N-octet packets per second
1326
1327 See Also:
1328 Classification
1329 Stream
1330 In-sequence Packet
1331 Intended Vector
1332 Offered Vector
1333 Expected Vector
1334
1335
1336
1337
1338
1339
1340 <span class="grey">Poretsky, et al. Informational [Page 24]</span></pre>
1341 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-25" ></span>
1342 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1343
1344
1345 <span class="h5"><a class="selflink" id="section-3.4.4.4" href="#section-3.4.4.4">3.4.4.4</a>. Instantaneous Delay Vector</span>
1346
1347 Definition:
1348 The instantaneous Forwarding Delay for a packet in a stream
1349 matching a specific classification, such as DSCP, that a DUT/SUT
1350 is measured to transmit to the correct destination interface
1351 successfully in response to an offered vector.
1352
1353 Discussion:
1354 Instantaneous Delay Vector is expressed as a combination of
1355 values: the classification rules AND Forwarding Delay. For every
1356 packet received in a Forwarding Vector, there is a corresponding
1357 Instantaneous Delay Vector.
1358
1359 Instantaneous Delay Vector is a per-hop measurement. The DUT/SUT
1360 MAY remark the specific DSCP or IP precedence value for a multi-
1361 hop measurement. The stream remains the same.
1362
1363 Instantaneous Delay Vector can be obtained at any offered load.
1364 It is RECOMMENDED that this vector be obtained at or below the
1365 Forwarding Capacity in the absence of Forwarding Congestion. For
1366 congested Forwarding Delay, run the offered load above the
1367 Forwarding Capacity.
1368
1369 Measurement Units:
1370 milliseconds
1371
1372 See Also:
1373 Classification
1374 Stream
1375 Forwarding Capacity
1376 Forwarding Delay
1377 Intended Vector
1378 Offered Vector
1379 Expected Delay Vector
1380
1381 <span class="h5"><a class="selflink" id="section-3.4.4.5" href="#section-3.4.4.5">3.4.4.5</a>. Average Delay Vector</span>
1382
1383 Definition:
1384 The average Forwarding Delay for packets in a stream matching a
1385 specific classification, such as DSCP, that a DUT/SUT is measured
1386 to transmit to the correct destination interface successfully in
1387 response to an offered vector.
1388
1389
1390
1391
1392
1393
1394
1395
1396 <span class="grey">Poretsky, et al. Informational [Page 25]</span></pre>
1397 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-26" ></span>
1398 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1399
1400
1401 Discussion:
1402 Average Delay Vector is expressed as combination of values: the
1403 classification rules AND average Forwarding Delay.
1404
1405 The average Forwarding Delay is computed by averaging all the
1406 Instantaneous Delay Vectors for a given stream.
1407
1408 Average Delay Vector is a per-hop measurement. The DUT/SUT MAY
1409 remark the specific DSCP or IP precedence value for a multi-hop
1410 measurement. The stream remains the same.
1411
1412 Average Delay Vector can be obtained at any offered load. It is
1413 recommended that the offered load be at or below the Forwarding
1414 Capacity in the absence of congestion. For congested Forwarding
1415 Delay, run the offered load above the Forwarding Capacity.
1416
1417 Measurement Units:
1418 milliseconds
1419
1420 See Also:
1421 Classification
1422 Stream
1423 Forwarding Capacity
1424 Forwarding Delay
1425 Intended Vector
1426 Offered Vector
1427 Expected Delay Vector
1428 Instantaneous Delay Vector
1429
1430 <span class="h5"><a class="selflink" id="section-3.4.4.6" href="#section-3.4.4.6">3.4.4.6</a>. Maximum Delay Vector</span>
1431
1432 Definition:
1433 The maximum Forwarding Delay for packets in a stream matching a
1434 specific classification, such as DSCP, that a DUT/SUT is measured
1435 to transmit to the correct destination interface successfully in
1436 response to an offered vector.
1437
1438 Discussion:
1439 Maximum Delay Vector is expressed as combination of values: the
1440 classification rules AND maximum Forwarding Delay.
1441
1442 The maximum Forwarding Delay is computed by selecting the highest
1443 value from the Instantaneous Delay Vectors for a given stream.
1444
1445 Maximum Delay Vector is a per-hop measurement. The DUT/SUT MAY
1446 remark the specific DSCP or IP precedence value for a multi-hop
1447 measurement. The stream remains the same.
1448
1449
1450
1451
1452 <span class="grey">Poretsky, et al. Informational [Page 26]</span></pre>
1453 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-27" ></span>
1454 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1455
1456
1457 Maximum Delay Vector can be obtained at any offered load. It is
1458 recommended that the offered load be at or below the Forwarding
1459 Capacity in the absence of congestion. For congested Forwarding
1460 Delay, run the offered load above the Forwarding Capacity.
1461
1462 Measurement Units:
1463 milliseconds
1464
1465 See Also:
1466 Classification
1467 Stream
1468 Forwarding Capacity
1469 Forwarding Delay
1470 Intended Vector
1471 Offered Vector
1472 Expected Delay Vector
1473 Instantaneous Delay Vector
1474
1475 <span class="h5"><a class="selflink" id="section-3.4.4.7" href="#section-3.4.4.7">3.4.4.7</a>. Minimum Delay Vector</span>
1476
1477 Definition:
1478 The minimum Forwarding Delay for packets in a stream matching a
1479 specific classification, such as DSCP, that a DUT/SUT is measured
1480 to transmit to the correct destination interface successfully in
1481 response to an offered vector.
1482
1483 Discussion:
1484 Minimum Delay Vector is expressed as a combination of values: the
1485 classification rules AND minimum Forwarding Delay. The minimum
1486 Forwarding Delay is computed by selecting the lowest value from
1487 the Instantaneous Delay Vectors for a given stream.
1488
1489 Minimum Delay Vector is a per-hop measurement. The DUT/SUT MAY
1490 remark the specific DSCP or IP precedence value for a multi-hop
1491 measurement. The stream remains the same.
1492
1493 Minimum Delay Vector can be obtained at any offered load. It is
1494 recommended that the offered load be at or below the Forwarding
1495 Capacity in the absence of congestion. For congested Forwarding
1496 Delay, run the offered load above the Forwarding Capacity.
1497
1498 Measurement Units:
1499 milliseconds
1500
1501
1502
1503
1504
1505
1506
1507
1508 <span class="grey">Poretsky, et al. Informational [Page 27]</span></pre>
1509 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-28" ></span>
1510 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1511
1512
1513 See Also:
1514 Classification
1515 Stream
1516 Forwarding Capacity
1517 Forwarding Delay
1518 Intended Vector
1519 Offered Vector
1520 Expected Delay Vector
1521
1522 <span class="h5"><a class="selflink" id="section-3.4.4.8" href="#section-3.4.4.8">3.4.4.8</a>. Instantaneous Jitter Vector</span>
1523
1524 Definition:
1525 The jitter for two consecutive packets in a stream matching a
1526 specific classification, such as DSCP, that a DUT/SUT is measured
1527 to transmit to the correct destination interface successfully in
1528 response to an offered vector.
1529
1530 Discussion:
1531 Instantaneous Jitter is the absolute value of the difference
1532 between the Forwarding Delay measurement of two packets belonging
1533 to the same stream.
1534
1535 The Instantaneous Jitter vector is expressed as a pair of numbers.
1536 Both the specific DSCP (or IP precedence) value AND jitter value
1537 combine to make a vector.
1538
1539 The Forwarding Delay fluctuation between two consecutive packets
1540 in a stream is reported as the "Instantaneous Jitter".
1541 Instantaneous Jitter Vector can be expressed as |D(i) - D(i-1)|,
1542 where D equals the Forwarding Delay and i is the test sequence
1543 number. Packets lost are not counted in the measurement.
1544
1545 The Instantaneous Jitter Vector is a per-hop measurement. The
1546 DUT/SUT MAY remark the specific DSCP or IP precedence value for a
1547 multi-hop measurement. The stream remains the same.
1548
1549 There may be several Instantaneous Jitter Vectors for a single
1550 stream. For n packets measured, there may be (n-1) Instantaneous
1551 Jitter Vectors.
1552
1553 Measurement units:
1554 milliseconds
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564 <span class="grey">Poretsky, et al. Informational [Page 28]</span></pre>
1565 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-29" ></span>
1566 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1567
1568
1569 See Also:
1570 Classification
1571 Stream
1572 Forwarding Delay
1573 Jitter
1574 Forwarding Vector
1575 Expected Vectors
1576
1577 <span class="h5"><a class="selflink" id="section-3.4.4.9" href="#section-3.4.4.9">3.4.4.9</a>. Average Jitter Vector</span>
1578
1579 Definition:
1580 The average jitter for packets in a stream matching a specific
1581 classification, such as DSCP, that a DUT/SUT is measured to
1582 transmit to the correct destination interface successfully in
1583 response to an offered vector.
1584
1585 Discussion:
1586 Average jitter is calculated by the average of all the
1587 Instantaneous Jitter Vectors of the same stream measured during
1588 the test duration. Average Jitter Vector is expressed as a
1589 combination of values: the classification rules AND average
1590 Jitter.
1591
1592 Average Jitter Vector is a per-hop measurement. The DUT/SUT MAY
1593 remark the specific DSCP or IP precedence value for a multi-hop
1594 measurement. The stream remains the same.
1595
1596 Measurement units:
1597 milliseconds
1598
1599 See Also:
1600 Classification
1601 Stream
1602 Jitter
1603 Forwarding Vector
1604 Expected Vector
1605 Instantaneous Jitter Vector
1606
1607 <span class="h5"><a class="selflink" id="section-3.4.4.10" href="#section-3.4.4.10">3.4.4.10</a>. Peak-to-peak Jitter Vector</span>
1608
1609 Definition:
1610 The maximum possible variation in the Forwarding Delay for packets
1611 in a stream matching a specific classification, such as DSCP, that
1612 a DUT/SUT is measured to transmit to the correct destination
1613 interface successfully in response to an offered vector.
1614
1615
1616
1617
1618
1619
1620 <span class="grey">Poretsky, et al. Informational [Page 29]</span></pre>
1621 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-30" ></span>
1622 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1623
1624
1625 Discussion:
1626 Peak-to-peak Jitter Vector is calculated by subtracting the
1627 maximum Forwarding Delay from the minimum Forwarding Delay of the
1628 packets forwarded by the DUT/SUT. Jitter vector is expressed as a
1629 combination of values: the classification rules AND peak-to-peak
1630 Jitter.
1631
1632 Peak-to-peak Jitter is not derived from the Instantaneous Jitter
1633 Vector. Peak-to-peak Jitter is based upon all the packets during
1634 the test duration, not just two consecutive packets.
1635
1636 Measurement units:
1637 milliseconds
1638
1639 See Also:
1640 Jitter
1641 Forwarding Vector
1642 Stream
1643 Expected Vectors
1644 Instantaneous Jitter Vector
1645 Average Jitter Vector
1646
1647 <span class="h2"><a class="selflink" id="section-4" href="#section-4">4</a>. Security Considerations</span>
1648
1649 Documents of this type do not directly affect the security of the
1650 Internet or of corporate networks as long as benchmarking is not
1651 performed on devices or systems connected to production networks.
1652
1653 Packets with unintended and/or unauthorized DSCP or IP precedence
1654 values may present security issues. Determining the security
1655 consequences of such packets is out of scope for this document.
1656
1657 <span class="h2"><a class="selflink" id="section-5" href="#section-5">5</a>. Acknowledgements</span>
1658
1659 The authors gratefully acknowledge the contributions of the IETF's
1660 Benchmarking Methodology Working Group members in reviewing this
1661 document. The authors would like to express our thanks to David
1662 Newman for his consistent and valuable assistance throughout the
1663 development of this document. The authors would also like to thank
1664 Al Morton and Kevin Dubray for their ideas and support.
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676 <span class="grey">Poretsky, et al. Informational [Page 30]</span></pre>
1677 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-31" ></span>
1678 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1679
1680
1681 <span class="h2"><a class="selflink" id="section-6" href="#section-6">6</a>. References</span>
1682
1683 <span class="h3"><a class="selflink" id="section-6.1" href="#section-6.1">6.1</a>. Normative References</span>
1684
1685 [<a id="ref-Br91">Br91</a>] Bradner, S., "Benchmarking terminology for network
1686 interconnection devices", <a href="./rfc1242">RFC 1242</a>, July 1991.
1687
1688 [<a id="ref-Br97">Br97</a>] Bradner, S., "Key words for use in RFCs to Indicate
1689 Requirement Levels", <a href="https://www.rfc-editor.org/bcp/bcp14">BCP 14</a>, <a href="./rfc2119">RFC 2119</a>, March 1997.
1690
1691 [<a id="ref-Br98">Br98</a>] Braden, B., Clark, D., Crowcroft, J., Davie, B., Deering, S.,
1692 Estrin, D., Floyd, S., Jacobson, V., Minshall, G., Partridge,
1693 C., Peterson, L., Ramakrishnan, K., Shenker, S., Wroclawski,
1694 J., and L. Zhang, "Recommendations on Queue Management and
1695 Congestion Avoidance in the Internet", <a href="./rfc2309">RFC 2309</a>, April 1998.
1696
1697 [<a id="ref-Ma98">Ma98</a>] Mandeville, R., "Benchmarking Terminology for LAN Switching
1698 Devices", <a href="./rfc2285">RFC 2285</a>, February 1998.
1699
1700 [<a id="ref-Ni98">Ni98</a>] Nichols, K., Blake, S., Baker, F., and D. Black, "Definition
1701 of the Differentiated Services Field (DS Field) in the IPv4
1702 and IPv6 Headers", <a href="./rfc2474">RFC 2474</a>, December 1998.
1703
1704 [<a id="ref-St91">St91</a>] Steinberg, L., "Techniques for managing asynchronously
1705 generated alerts", <a href="./rfc1224">RFC 1224</a>, May 1991.
1706
1707 <span class="h3"><a class="selflink" id="section-6.2" href="#section-6.2">6.2</a>. Informative References</span>
1708
1709 [<a id="ref-Al99">Al99</a>] Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way Delay
1710 Metric for IPPM", <a href="./rfc2679">RFC 2679</a>, September 1999.
1711
1712 [<a id="ref-Bl98">Bl98</a>] Blake, S., Black, D., Carlson, M., Davies, E., Wang, Z., and
1713 W. Weiss, "An Architecture for Differentiated Service", <a href="./rfc2475">RFC</a>
1714 <a href="./rfc2475">2475</a>, December 1998.
1715
1716 [<a id="ref-Br99">Br99</a>] Bradner, S. and J. McQuaid, "Benchmarking Methodology for
1717 Network Interconnect Devices", <a href="./rfc2544">RFC 2544</a>, March 1999.
1718
1719 [<a id="ref-De02">De02</a>] Demichelis, C. and P. Chimento, "IP Packet Delay Variation
1720 Metric for IP Performance Metrics (IPPM)", <a href="./rfc3393">RFC 3393</a>, November
1721 2002.
1722
1723 [<a id="ref-Ec98">Ec98</a>] <a href="http://www3.ietf.org/proceedings/98mar/98mar-edited-135.htm">http://www3.ietf.org/proceedings/98mar/98mar-edited-135.htm</a>
1724
1725 [<a id="ref-Fl93">Fl93</a>] Floyd, S., and Jacobson, V., "Random Early Detection gateways
1726 for Congestion Avoidance", IEEE/ACM Transactions on
1727 Networking, V.1 N.4, August 1993, p. 397-413. URL
1728 "<a href="ftp://ftp.ee.lbl.gov/papers/early.pdf">ftp://ftp.ee.lbl.gov/papers/early.pdf</a>".
1729
1730
1731
1732 <span class="grey">Poretsky, et al. Informational [Page 31]</span></pre>
1733 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-32" ></span>
1734 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1735
1736
1737 [<a id="ref-Ja99">Ja99</a>] Davie, B., Charny, A., Bennet, J.C., Benson, K., Le Boudec,
1738 J., Courtney, W., Davari, S., Firoiu, V., and D. Stiliadis,
1739 "An Expedited Forwarding PHB (Per-Hop Behavior)", <a href="./rfc3246">RFC 3246</a>,
1740 March 2002.
1741
1742 [<a id="ref-Ka99">Ka99</a>] Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way Packet
1743 Loss Metric for IPPM", <a href="./rfc2680">RFC 2680</a>, September 1999.
1744
1745 [<a id="ref-Ma91">Ma91</a>] Mankin, A. and K. Ramakrishnan, "Gateway Congestion Control
1746 Survey", <a href="./rfc1254">RFC 1254</a>, August 1991.
1747
1748 [<a id="ref-Ma00">Ma00</a>] Mandeville, R. and J. Perser, "Benchmarking Methodology for
1749 LAN Switching Devices", <a href="./rfc2889">RFC 2889</a>, August 2000.
1750
1751 [<a id="ref-Mo03">Mo03</a>] Morton, A., Ciavattone, L., Ramachandran, G., Shalunov, S.,
1752 Perser, J., "Packet Reordering Metric for IPPM", Work in
1753 Progress.
1754
1755 [<a id="ref-Na84">Na84</a>] Nagle, J., "Congestion control in IP/TCP internetworks", <a href="./rfc896">RFC</a>
1756 <a href="./rfc896">896</a>, January 1984.
1757
1758 [<a id="ref-Ra99">Ra99</a>] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition of
1759 Explicit Congestion Notification (ECN) to IP", <a href="./rfc3168">RFC 3168</a>,
1760 September 2001.
1761
1762 [<a id="ref-Sc96">Sc96</a>] Schulzrinne, H., Casner, S., Frederick, R., and V. Jacobson,
1763 "RTP: A Transport Protocol for Real-Time Applications", STD
1764 64, <a href="./rfc3550">RFC 3550</a>, July 2003.
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788 <span class="grey">Poretsky, et al. Informational [Page 32]</span></pre>
1789 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-33" ></span>
1790 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1791
1792
1793 Authors' Addresses
1794
1795 Jerry Perser
1796 Veriwave
1797 8770 SW Nimbus Ave.
1798 Suite B
1799 Beaverton, OR 97008 USA
1800 USA
1801
1802 Phone: + 1 818 338 4112
1803 EMail: [email protected]
1804
1805
1806 Scott Poretsky
1807 Reef Point Systems
1808 8 New England Executive Park
1809 Burlington, MA 01803
1810 USA
1811
1812 Phone: + 1 508 439 9008
1813 EMail: [email protected]
1814
1815
1816 Shobha Erramilli
1817 Telcordia Technologies
1818 331 Newman Springs Road
1819 Red Bank, New Jersey 07701
1820 USA
1821
1822 EMail: [email protected]
1823
1824
1825 Sumit Khurana
1826 Motorola
1827 7700 West Parmer Ln.
1828 Austin, TX 78729
1829 USA
1830
1831 Phone: +1 512 996 6604
1832 Email: [email protected]
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844 <span class="grey">Poretsky, et al. Informational [Page 33]</span></pre>
1845 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-34" ></span>
1846 <span class="grey"><a href="./rfc4689">RFC 4689</a> Terminology for Traffic Control Mechanisms October 2006</span>
1847
1848
1849 Full Copyright Statement
1850
1851 Copyright (C) The Internet Society (2006).
1852
1853 This document is subject to the rights, licenses and restrictions
1854 contained in <a href="https://www.rfc-editor.org/bcp/bcp78">BCP 78</a>, and except as set forth therein, the authors
1855 retain all their rights.
1856
1857 This document and the information contained herein are provided on an
1858 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
1859 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
1860 ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED,
1861 INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
1862 INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
1863 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
1864
1865 Intellectual Property
1866
1867 The IETF takes no position regarding the validity or scope of any
1868 Intellectual Property Rights or other rights that might be claimed to
1869 pertain to the implementation or use of the technology described in
1870 this document or the extent to which any license under such rights
1871 might or might not be available; nor does it represent that it has
1872 made any independent effort to identify any such rights. Information
1873 on the procedures with respect to rights in RFC documents can be
1874 found in <a href="https://www.rfc-editor.org/bcp/bcp78">BCP 78</a> and <a href="https://www.rfc-editor.org/bcp/bcp79">BCP 79</a>.
1875
1876 Copies of IPR disclosures made to the IETF Secretariat and any
1877 assurances of licenses to be made available, or the result of an
1878 attempt made to obtain a general license or permission for the use of
1879 such proprietary rights by implementers or users of this
1880 specification can be obtained from the IETF on-line IPR repository at
1881 <a href="http://www.ietf.org/ipr">http://www.ietf.org/ipr</a>.
1882
1883 The IETF invites any interested party to bring to its attention any
1884 copyrights, patents or patent applications, or other proprietary
1885 rights that may cover technology that may be required to implement
1886 this standard. Please address the information to the IETF at
1887 [email protected].
1888
1889 Acknowledgement
1890
1891 Funding for the RFC Editor function is provided by the IETF
1892 Administrative Support Activity (IASA).
1893
1894
1895
1896
1897
1898
1899
1900 Poretsky, et al. Informational [Page 34]
1901 </pre>