idnits 2.17.1 draft-ietf-tcpm-rtorestart-06.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (April 8, 2015) is 3296 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- == Missing Reference: 'SEG 1' is mentioned on line 166, but not defined == Missing Reference: 'SEG 2' is mentioned on line 167, but not defined == Missing Reference: 'SEG 3' is mentioned on line 172, but not defined ** Obsolete normative reference: RFC 4960 (Obsoleted by RFC 9260) Summary: 1 error (**), 0 flaws (~~), 4 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 TCP Maintenance and Minor Extensions (tcpm) P. Hurtig 3 Internet-Draft A. Brunstrom 4 Intended status: Experimental Karlstad University 5 Expires: October 10, 2015 A. Petlund 6 Simula Research Laboratory AS 7 M. Welzl 8 University of Oslo 9 April 8, 2015 11 TCP and SCTP RTO Restart 12 draft-ietf-tcpm-rtorestart-06 14 Abstract 16 This document describes a modified algorithm for managing the TCP and 17 SCTP retransmission timers that provides faster loss recovery when 18 there is a small amount of outstanding data for a connection. The 19 modification, RTO Restart (RTOR), allows the transport to restart its 20 retransmission timer more aggressively in situations where fast 21 retransmit cannot be used. This enables faster loss detection and 22 recovery for connections that are short-lived or application-limited. 24 Status of This Memo 26 This Internet-Draft is submitted in full conformance with the 27 provisions of BCP 78 and BCP 79. 29 Internet-Drafts are working documents of the Internet Engineering 30 Task Force (IETF). Note that other groups may also distribute 31 working documents as Internet-Drafts. The list of current Internet- 32 Drafts is at http://datatracker.ietf.org/drafts/current/. 34 Internet-Drafts are draft documents valid for a maximum of six months 35 and may be updated, replaced, or obsoleted by other documents at any 36 time. It is inappropriate to use Internet-Drafts as reference 37 material or to cite them other than as "work in progress." 39 This Internet-Draft will expire on October 10, 2015. 41 Copyright Notice 43 Copyright (c) 2015 IETF Trust and the persons identified as the 44 document authors. All rights reserved. 46 This document is subject to BCP 78 and the IETF Trust's Legal 47 Provisions Relating to IETF Documents 48 (http://trustee.ietf.org/license-info) in effect on the date of 49 publication of this document. Please review these documents 50 carefully, as they describe your rights and restrictions with respect 51 to this document. Code Components extracted from this document must 52 include Simplified BSD License text as described in Section 4.e of 53 the Trust Legal Provisions and are provided without warranty as 54 described in the Simplified BSD License. 56 1. Introduction 58 TCP uses two mechanisms to detect segment loss. First, if a segment 59 is not acknowledged within a certain amount of time, a retransmission 60 timeout (RTO) occurs, and the segment is retransmitted [RFC6298]. 61 While the RTO is based on measured round-trip times (RTTs) between 62 the sender and receiver, it also has a conservative lower bound of 1 63 second to ensure that delayed segments are not mistaken as lost. 64 Second, when a sender receives dupACKs, the fast retransmit algorithm 65 infers segment loss and triggers a retransmission [RFC5681]. 66 Duplicate acknowledgments are generated by a receiver when out-of- 67 order segments arrive. As both segment loss and segment reordering 68 cause out-of-order arrival, fast retransmit waits for three dupACKs 69 before considering the segment as lost. In some situations, however, 70 the number of outstanding segments is not enough to trigger three 71 dupACKs, and the sender must rely on lengthy RTOs for loss recovery. 73 The number of outstanding segments can be small for several reasons: 75 (1) The connection is limited by the congestion control when the 76 path has a low total capacity (bandwidth-delay product) or the 77 connection's share of the capacity is small. It is also limited 78 by the congestion control in the first few RTTs of a connection 79 or after an RTO when the available capacity is probed using 80 slow-start. 82 (2) The connection is limited by the receiver's available buffer 83 space. 85 (3) The connection is limited by the application if the available 86 capacity of the path is not fully utilized (e.g. interactive 87 applications), or at the end of a transfer. 89 While the reasons listed above are valid for any flow, the third 90 reason is most common for applications that transmit short flows, or 91 use a bursty transmission pattern. A typical example of applications 92 that produce short flows are web-based applications. [RJ10] shows 93 that 70% of all web objects, found at the top 500 sites, are too 94 small for fast retransmit to work. [FDT13] shows that about 77% of 95 all retransmissions sent by a major web service are sent after RTO 96 expiry. Applications with bursty transmission patterns often send 97 data in response to actions, or as a reaction to real life events. 98 Typical examples of such applications are stock trading systems, 99 remote computer operations, online games, and web-based applications 100 using persistent connections. What is special about this class of 101 applications is that they often are time-dependant, and extra latency 102 can reduce the application service level [P09]. 104 The RTO Restart (RTOR) mechanism described in this document makes the 105 RTO slightly more aggressive when the number of outstanding segments 106 is too small for fast retransmit to work, in an attempt to enable 107 faster loss recovery for all segments while being robust to 108 reordering. While RTOR still conforms to the requirement in 109 [RFC6298] that segments must not be retransmitted earlier than RTO 110 seconds after their original transmission, it could increase the risk 111 of spurious timeout. Spurious timeouts can degrade the performance 112 of flows with multiple bursts of data, as a burst following a 113 spurious timeout might not fit within the reduced congestion window 114 (cwnd). There are, however, several techniques to mitigate the 115 effects of such unnecessary retransmissions (cf. [RFC4015]). 117 While this document focuses on TCP, the described changes are also 118 valid for the Stream Control Transmission Protocol (SCTP) [RFC4960] 119 which has similar loss recovery and congestion control algorithms. 121 2. Terminology 123 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 124 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 125 document are to be interpreted as described in RFC 2119 [RFC2119]. 127 This document introduces the following variables: 129 The number of previously unsent segments (prevunsnt): The number of 130 segments that a sender has queued for transmission, but has not yet 131 sent. 133 RTO Restart threshold (rrthresh): RTOR is enabled whenever the sum of 134 the number of outstanding and previously unsent segments (prevunsnt) 135 is below this threshold. 137 3. RTO Restart Overview 139 The RTO management algorithm described in [RFC6298] recommends that 140 the retransmission timer is restarted when an acknowledgment (ACK) 141 that acknowledges new data is received and there is still outstanding 142 data. The restart is conducted to guarantee that unacknowledged 143 segments will be retransmitted after approximately RTO seconds. 144 However, by restarting the timer on each incoming ACK, 145 retransmissions are not typically triggered RTO seconds after their 146 previous transmission but rather RTO seconds after the last ACK 147 arrived. The duration of this extra delay depends on several factors 148 but is in most cases approximately one RTT. Hence, in most 149 situations, the time before a retransmission is triggered is equal to 150 "RTO + RTT". 152 The standardized RTO timer management is illustrated in Figure 1 153 where a TCP sender transmits three segments to a receiver. The 154 arrival of the first and second segment triggers a delayed ACK 155 (delACK) [RFC1122], which restarts the RTO timer at the sender. The 156 RTO is restarted approximately one RTT after the transmission of the 157 third segment. Thus, if the third segment is lost, as indicated in 158 Figure 1, the effective loss detection time is "RTO + RTT" seconds. 159 In some situations, the effective loss detection time becomes even 160 longer. Consider a scenario where only two segments are outstanding. 161 If the second segment is lost, the time to expire the delACK timer 162 will also be included in the effective loss detection time. 164 Sender Receiver 165 ... 166 DATA [SEG 1] ----------------------> (ack delayed) 167 DATA [SEG 2] ----------------------> (send ack) 168 DATA [SEG 3] ----X /-------- ACK 169 (restart RTO) <----------/ 170 ... 171 (RTO expiry) 172 DATA [SEG 3] ----------------------> 174 Figure 1: RTO restart example 176 During normal TCP bulk transfer the current RTO restart approach is 177 not a problem. Actually, as long as enough segments arrive at a 178 receiver to enable fast retransmit, RTO-based loss recovery should be 179 avoided. RTOs should only be used as a last resort, as they 180 drastically lower the congestion window compared to fast retransmit. 181 The current approach can therefore be beneficial -- it is described 182 in [EL04] to act as a "safety margin" that compensates for some of 183 the problems that the authors have identified with the standard RTO 184 calculation. Notably, the authors of [EL04] also state that "this 185 safety margin does not exist for highly interactive applications 186 where often only a single packet is in flight." 188 Although fast retransmit is preferrable there are situations where 189 timeouts are appropriate, or the only choice. For example, if the 190 network is severely congested and no segments arrive RTO-based 191 recovery should be used. In this situation, the time to recover from 192 the loss(es) will not be the performance bottleneck. However, for 193 connections that do not utilize enough capacity to enable fast 194 retransmit, RTO-based loss detection is the only choice and the time 195 required for this can become a performance bottleneck. 197 4. RTOR Algorithm 199 To enable faster loss recovery for connections that are unable to use 200 fast retransmit, RTOR can be used. By resetting the timer to "RTO - 201 T_earliest", where T_earliest is the time elapsed since the earliest 202 outstanding segment was transmitted, retransmissions will always 203 occur after exactly RTO seconds. This approach makes the RTO more 204 aggressive than the standardized approach in [RFC6298] but still 205 conforms to the requirement in [RFC6298] that segments must not be 206 retransmitted earlier than RTO seconds after their original 207 transmission. 209 This document specifies an OPTIONAL sender-only modification to TCP 210 and SCTP which updates step 5.3 in Section 5 of [RFC6298] (and a 211 similar update in Section 6.3.2 of [RFC4960] for SCTP). A sender 212 that implements this method MUST follow the algorithm below: 214 When an ACK is received that acknowledges new data: 216 (1) Set T_earliest = 0. 218 (2) If the sum of the number of outstanding and previously unsent 219 segments (prevunsnt) is less than an RTOR threshold 220 (rrthresh), set T_earliest to the time elapsed since the 221 earliest outstanding segment was sent. 223 (3) Restart the retransmission timer so that it will expire after 224 (for the current value of RTO): 226 (a) RTO - T_earliest, if RTO - T_earliest > 0. 228 (b) RTO, otherwise. 230 The RECOMMENDED value of rrthresh is four, as it will prevent RTOR 231 from being more aggressive and potentially causing RTOs instead of 232 fast retransmits. This update needs TCP implementations to track the 233 time elapsed since the transmission of the earliest outstanding 234 segment (T_earliest). As RTOR is only used when the amount of 235 outstanding and previously unsent data is less than rrthresh 236 segments, TCP implementations also need to track whether the amount 237 of outstanding and previously unsent data is more, equal, or less 238 than rrthresh segments. Although some packet-based TCP 239 implementations (e.g. Linux TCP) already track both the transmission 240 times of all segments and also the number of outstanding segments, 241 not all implementations do. Section 5.3 describes how to implement 242 segment tracking for a general TCP implementation. To use RTOR, the 243 calculated expiration time MUST be positive (step 3(a) in the list 244 above); this is required to ensure that RTOR does not trigger 245 retransmissions prematurely when previously retransmitted segments 246 are acknowledged. 248 5. Discussion 250 In this section, we discuss the applicability and a number of issues 251 surrounding RTOR. 253 5.1. Applicability 255 The currently standardized algorithm has been shown to add at least 256 one RTT to the loss recovery process in TCP [LS00] and SCTP 257 [HB11][PBP09]. For applications that have strict timing requirements 258 (e.g. interactive web) rather than throughput requirements, using 259 RTOR could be beneficial because the RTT and also the delACK timer of 260 receivers are often large components of the effective loss recovery 261 time. Measurements in [HB11] have shown that the total transfer time 262 of a lost segment (including the original transmission time and the 263 loss recovery time) can be reduced by 35% using RTOR. These results 264 match those presented in [PGH06][PBP09], where RTOR is shown to 265 significantly reduce retransmission latency. 267 There are also traffic types that do not benefit from RTOR. One 268 example of such traffic is bulk transmission. The reason why bulk 269 traffic does not benefit from RTOR is that such traffic flows mostly 270 have four or more segments outstanding, allowing loss recovery by 271 fast retransmit. However, there is no harm in using RTOR for such 272 traffic as the algorithm only is active when the amount of 273 outstanding and unsent segments are less than rrthresh (default 4). 275 Given that RTOR is a mostly conservative algorithm, it is suitable 276 for experimentation as a system-wide default for TCP traffic. 278 5.2. Spurious Timeouts 280 RTOR can in some situations reduce the loss detection time and 281 thereby increase the risk of spurious timeouts. In theory, the 282 retransmission timer has a lower bound of 1 second [RFC6298], which 283 limits the risk of having spurious timeouts. However, in practice 284 most implementations use a significantly lower value. Initial 285 measurements, show slight increases in the number of spurious 286 timeouts when such lower values are used [RHB15]. However, further 287 experiments, in different environments and with different types of 288 traffic, are encouraged to quantify such increases more reliably. 290 Does a slightly increased risk matter? Generally, spurious timeouts 291 have a negative effect on the network as segments are transmitted 292 needlessly. However, recent experiments do not show a significant 293 increase in network load for a number of realistic scenarios [RHB15]. 294 Another problem with spurious retransmissions is related to the 295 performance of TCP/SCTP, as the congestion window is reduced to one 296 segment when timeouts occur [RFC5681]. This could be a potential 297 problem for applications transmitting multiple bursts of data within 298 a single flow, e.g. web-based HTTP/1.1 and HTTP/2.0 applications. 299 However, results from recent experiments involving persistent web 300 traffic [RHB15] only revealed a net gain of using RTOR. Other types 301 of flows, e.g. long-lived bulk flows, are not affected as the 302 algorithm is only applied when the amount of outstanding and unsent 303 segments is less than rrthresh. Furthermore, short-lived and 304 application-limited flows are typically not affected as they are too 305 short to experience the effect of congestion control or have a 306 transmission rate that is quickly attainable. 308 While a slight increase in spurious timeouts has been observed using 309 RTOR, it is not clear whether the effects of this increase mandate 310 any future algorithmic changes or not -- especially since most modern 311 operating systems already include mechanisms to detect 312 [RFC3522][RFC3708][RFC5682] and resolve [RFC4015] possible problems 313 with spurious retransmissions. Further experimentation is needed to 314 determine this and thereby move this specification from experimental 315 to proposed standard. For instance, RTOR has not been evaluated in 316 the context of mobile networks. Mobile networks often incur highly 317 variable RTTs (delay spikes), due to e.g. handovers, and would 318 therefore be a useful scenario for further experimentation. 320 5.3. Tracking Outstanding and Previously Unsent Segments 322 The method of tracking outstanding and previously unsent segments 323 will probably differ depending on the actual TCP implementation. For 324 packet-based TCP implementations, tracking outstanding segments is 325 often straightforward and can be implemented using a simple counter. 326 For byte-based TCP stacks it is a more complex task. Section 3.2 of 327 [RFC5827] outlines a general method of tracking the number of 328 outstanding segments. The same method can be used for RTOR. The 329 implementation will have to track segment boundaries to form an 330 understanding as to how many actual segments have been transmitted, 331 but not acknowledged. This can be done by the sender tracking the 332 boundaries of the rrthresh segments on the right side of the current 333 window (which involves tracking rrthresh + 1 sequence numbers in 334 TCP). This could be done by keeping a circular list of the segment 335 boundaries, for instance. Cumulative ACKs that do not fall within 336 this region indicate that at least rrthresh segments are outstanding, 337 and therefore RTOR is not enabled. When the outstanding window 338 becomes small enough that RTOR can be invoked, a full understanding 339 of the number of outstanding segments will be available from the 340 rrthresh + 1 sequence numbers retained. (Note: the implicit sequence 341 number consumed by the TCP FIN bit can also be included in the 342 tracking of segment boundaries.) 344 Tracking the number of previously unsent segments depends on the 345 segmentation strategy used by the TCP implementation, not whether it 346 is packet-based or byte-based. In the case segments are formed 347 directly on socket writes, the process of determining the number of 348 previously unsent segments should be trivial. In the case that 349 unsent data can be segmented (or re-segmented) as long as it still is 350 unsent, a straightforward strategy could be to divide the amount of 351 unsent data (in bytes) with the SMSS to obtain an estimate. In some 352 cases, such an estimation could be too simplistic, depending on the 353 segmentation strategy of the TCP implementation. However, this 354 estimation is not critical to RTOR. For instance, implementations 355 can use a simplified method by setting prevunsnt to rrthresh whenever 356 previously unsent data is available, and set prevunsnt to zero when 357 no new data is available. This will disable RTOR in the presence of 358 unsent data and only use the number of outstanding segments to 359 enable/disable RTOR. This strategy was used in an earlier version of 360 the algorithm and works well. The addition of tracking prevunsnt was 361 only made to optimize a corner case in which RTOR was unnecessarily 362 disabled. 364 6. Related Work 366 There are several proposals that address the problem of not having 367 enough ACKs for loss recovery. In what follows, we explain why the 368 mechanism described here is complementary to these approaches: 370 The limited transmit mechanism [RFC3042] allows a TCP sender to 371 transmit a previously unsent segment for each of the first two 372 dupACKs. By transmitting new segments, the sender attempts to 373 generate additional dupACKs to enable fast retransmit. However, 374 limited transmit does not help if no previously unsent data is ready 375 for transmission. [RFC5827] specifies an early retransmit algorithm 376 to enable fast loss recovery in such situations. By dynamically 377 lowering the number of dupACKs needed for fast retransmit 378 (dupthresh), based on the number of outstanding segments, a smaller 379 number of dupACKs is needed to trigger a retransmission. In some 380 situations, however, the algorithm is of no use or might not work 381 properly. First, if a single segment is outstanding, and lost, it is 382 impossible to use early retransmit. Second, if ACKs are lost, early 383 retransmit cannot help. Third, if the network path reorders 384 segments, the algorithm might cause more unnecessary retransmissions 385 than fast retransmit. The recommended value of RTOR's rrthresh 386 variable is based on the dupthresh, but is possible to adapt to allow 387 tighter integration with other experimental algorithms such as early 388 retransmit. 390 Tail Loss Probe [TLP] is a proposal to send up to two "probe 391 segments" when a timer fires which is set to a value smaller than the 392 RTO. A "probe segment" is a new segment if new data is available, 393 else a retransmission. The intention is to compensate for sluggish 394 RTO behavior in situations where the RTO greatly exceeds the RTT, 395 which, according to measurements reported in [TLP], is not uncommon. 396 Furthermore, TLP also tries to circumvent the congestion window reset 397 to one segment by instead enabling fast recovery. The Probe timeout 398 (PTO) is normally two RTTs, and a spurious PTO is less risky than a 399 spurious RTO because it would not have the same negative effects 400 (clearing the scoreboard and restarting with slow-start). TLP is a 401 more advanced mechanism than RTOR, requiring e.g. SACK to work, and 402 is often able to reduce loss recovery times more. However, it also 403 increases the amount of spurious retransmissions noticeably, as 404 compared to RTOR [RHB15]. 406 TLP is applicable in situations where RTOR does not apply, and it 407 could overrule (yielding a similar general behavior, but with a lower 408 timeout) RTOR in cases where the number of outstanding segments is 409 smaller than four and no new segments are available for transmission. 410 The PTO has the same inherent problem of restarting the timer on an 411 incoming ACK, and could be combined with a strategy similar to RTOR's 412 to offer more consistent timeouts. 414 7. SCTP Socket API Considerations 416 This section describes how the socket API for SCTP defined in 417 [RFC6458] is extended to control the usage of RTO restart for SCTP. 419 Please note that this section is informational only. 421 7.1. Data Types 423 This section uses data types from [IEEE.1003-1G.1997]: uintN_t means 424 an unsigned integer of exactly N bits (e.g., uint16_t). This is the 425 same as in [RFC6458]. 427 7.2. Socket Option for Controlling the RTO Restart Support 428 (SCTP_RTO_RESTART) 430 This socket option allows the enabling or disabling of RTO Restart 431 for SCTP associations. 433 Whether RTO Restart is enabled or not per default is implementation 434 specific. 436 This socket option uses IPPROTO_SCTP as its level and 437 SCTP_RTO_RESTART as its name. It can be used with getsockopt() and 438 setsockopt(). The socket option value uses the following structure 439 defined in [RFC6458]: 441 struct sctp_assoc_value { 442 sctp_assoc_t assoc_id; 443 uint32_t assoc_value; 444 }; 446 assoc_id: This parameter is ignored for one-to-one style sockets. 447 For one-to-many style sockets, this parameter indicates upon which 448 association the user is performing an action. The special 449 sctp_assoc_t SCTP_{FUTURE|CURRENT|ALL}_ASSOC can also be used in 450 assoc_id for setsockopt(). For getsockopt(), the special value 451 SCTP_FUTURE_ASSOC can be used in assoc_id, but it is an error to 452 use SCTP_{CURRENT|ALL}_ASSOC in assoc_id. 454 assoc_value: A non-zero value encodes the enabling of RTO restart 455 whereas a value of 0 encodes the disabling of RTO restart. 457 sctp_opt_info() needs to be extended to support SCTP_RTO_RESTART. 459 8. IANA Considerations 461 This memo includes no request to IANA. 463 9. Security Considerations 465 This document discusses a change in how to set the retransmission 466 timer's value when restarted. Therefore, the security considerations 467 found in [RFC6298] apply to this document. No additional security 468 problems have been identified with RTO Restart at this time. 470 10. Acknowledgements 472 The authors wish to thank Michael Tuexen for contributing the SCTP 473 Socket API considerations and Godred Fairhurst, Yuchung Cheng, Mark 474 Allman, Anantha Ramaiah, Richard Scheffenegger, Nicolas Kuhn, 475 Alexander Zimmermann, and Michael Scharf for commenting on the draft 476 and the ideas behind it. 478 All the authors are supported by RITE (http://riteproject.eu/ ), a 479 research project (ICT-317700) funded by the European Community under 480 its Seventh Framework Program. The views expressed here are those of 481 the author(s) only. The European Commission is not liable for any 482 use that may be made of the information in this document. 484 11. Changes from Previous Versions 486 RFC-Editor note: please remove this section prior to publication. 488 11.1. Changes from draft-ietf-...-05 to -06 490 o Added socket API considerations, after discussing the draft in 491 tsvwg. 493 11.2. Changes from draft-ietf-...-04 to -05 495 o Introduced variable to track the number of previously unsent 496 segments. 498 o Clarified many concepts, e.g. extended the description on how to 499 track outstanding and previously unsent segments. 501 o Added a reference to initial measurements on the effects of using 502 RTOR. 504 o Improved wording throughout the document. 506 11.3. Changes from draft-ietf-...-03 to -04 508 o Changed the algorithm to allow RTOR when there is unsent data 509 available, but the cwnd does not allow transmission. 511 o Changed the algorithm to not trigger if RTOR <= 0. 513 o Made minor adjustments throughout the document to adjust for the 514 algorithmic change. 516 o Improved the wording throughout the document. 518 11.4. Changes from draft-ietf-...-02 to -03 520 o Updated the document to use "RTOR" instead of "RTO Restart" when 521 refering to the modified algorithm. 523 o Moved document terminology to a section of its own. 525 o Introduced the rrthresh variable in the terminology section. 527 o Added a section to generalize the tracking of outstanding 528 segments. 530 o Updated the algorithm to work when the number of outstanding 531 segments is less than four and one segment is ready for 532 transmission, by restarting the timer when new data has been sent. 534 o Clarified the relationship between fast retransmit and RTOR. 536 o Improved the wording throughout the document. 538 11.5. Changes from draft-ietf-...-01 to -02 540 o Changed the algorithm description in Section 3 to use formal RFC 541 2119 language. 543 o Changed last paragraph of Section 3 to clarify why the RTO restart 544 algorithm is active when less than four segments are outstanding. 546 o Added two paragraphs in Section 4.1 to clarify why the algorithm 547 can be turned on for all TCP traffic without having any negative 548 effects on traffic patterns that do not benefit from a modified 549 timer restart. 551 o Improved the wording throughout the document. 553 o Replaced and updated some references. 555 11.6. Changes from draft-ietf-...-00 to -01 557 o Improved the wording throughout the document. 559 o Removed the possibility for a connection limited by the receiver's 560 advertised window to use RTO restart, decreasing the risk of 561 spurious retransmission timeouts. 563 o Added a section that discusses the applicability of and problems 564 related to the RTO restart mechanism. 566 o Updated the text describing the relationship to TLP to reflect 567 updates made in this draft. 569 o Added acknowledgments. 571 12. References 573 12.1. Normative References 575 [RFC1122] Braden, R., "Requirements for Internet Hosts - 576 Communication Layers", STD 3, RFC 1122, October 1989. 578 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 579 Requirement Levels", BCP 14, RFC 2119, March 1997. 581 [RFC3042] Allman, M., Balakrishnan, H., and S. Floyd, "Enhancing 582 TCP's Loss Recovery Using Limited Transmit", RFC 3042, 583 January 2001. 585 [RFC3522] Ludwig, R. and M. Meyer, "The Eifel Detection Algorithm 586 for TCP", RFC 3522, April 2003. 588 [RFC3708] Blanton, E. and M. Allman, "Using TCP Duplicate Selective 589 Acknowledgement (DSACKs) and Stream Control Transmission 590 Protocol (SCTP) Duplicate Transmission Sequence Numbers 591 (TSNs) to Detect Spurious Retransmissions", RFC 3708, 592 February 2004. 594 [RFC4015] Ludwig, R. and A. Gurtov, "The Eifel Response Algorithm 595 for TCP", RFC 4015, February 2005. 597 [RFC4960] Stewart, R., "Stream Control Transmission Protocol", RFC 598 4960, September 2007. 600 [RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion 601 Control", RFC 5681, September 2009. 603 [RFC5682] Sarolahti, P., Kojo, M., Yamamoto, K., and M. Hata, 604 "Forward RTO-Recovery (F-RTO): An Algorithm for Detecting 605 Spurious Retransmission Timeouts with TCP", RFC 5682, 606 September 2009. 608 [RFC5827] Allman, M., Avrachenkov, K., Ayesta, U., Blanton, J., and 609 P. Hurtig, "Early Retransmit for TCP and Stream Control 610 Transmission Protocol (SCTP)", RFC 5827, May 2010. 612 [RFC6298] Paxson, V., Allman, M., Chu, J., and M. Sargent, 613 "Computing TCP's Retransmission Timer", RFC 6298, June 614 2011. 616 12.2. Informative References 618 [EL04] Ekstroem, H. and R. Ludwig, "The Peak-Hopper: A New End- 619 to-End Retransmission Timer for Reliable Unicast 620 Transport", IEEE INFOCOM 2004, March 2004. 622 [FDT13] Flach, T., Dukkipati, N., Terzis, A., Raghavan, B., 623 Cardwell, N., Cheng, Y., Jain, A., Hao, S., Katz-Bassett, 624 E., and R. Govindan, "Reducing Web Latency: the Virtue of 625 Gentle Aggression", Proc. ACM SIGCOMM Conf., August 2013. 627 [HB11] Hurtig, P. and A. Brunstrom, "SCTP: designed for timely 628 message delivery?", Springer Telecommunication Systems 47 629 (3-4), August 2011. 631 [IEEE.1003-1G.1997] 632 Institute of Electrical and Electronics Engineers, 633 "Protocol Independent Interfaces", IEEE Standard 1003.1G, 634 March 1997. 636 [LS00] Ludwig, R. and K. Sklower, "The Eifel retransmission 637 timer", ACM SIGCOMM Comput. Commun. Rev., 30(3), July 638 2000. 640 [P09] Petlund, A., "Improving latency for interactive, thin- 641 stream applications over reliable transport", Unipub PhD 642 Thesis, Oct 2009. 644 [PBP09] Petlund, A., Beskow, P., Pedersen, J., Paaby, E., Griwodz, 645 C., and P. Halvorsen, "Improving SCTP Retransmission 646 Delays for Time-Dependent Thin Streams", Springer 647 Multimedia Tools and Applications, 45(1-3), 2009. 649 [PGH06] Pedersen, J., Griwodz, C., and P. Halvorsen, 650 "Considerations of SCTP Retransmission Delays for Thin 651 Streams", IEEE LCN 2006, November 2006. 653 [RFC6458] Stewart, R., Tuexen, M., Poon, K., Lei, P., and V. 654 Yasevich, "Sockets API Extensions for the Stream Control 655 Transmission Protocol (SCTP)", RFC 6458, December 2011. 657 [RHB15] Rajiullah, M., Hurtig, P., Brunstrom, A., Petlund, A., and 658 M. Welzl, "An Evaluation of Tail Loss Recovery Mechanisms 659 for TCP", ACM SIGCOMM CCR 45 (1), January 2015. 661 [RJ10] Ramachandran, S., "Web metrics: Size and number of 662 resources", Google 663 http://code.google.com/speed/articles/web-metrics.html, 664 May 2010. 666 [TLP] Dukkipati, N., Cardwell, N., Cheng, Y., and M. Mathis, 667 "TCP Loss Probe (TLP): An Algorithm for Fast Recovery of 668 Tail Losses", Internet-draft draft-dukkipati-tcpm-tcp- 669 loss-probe-01.txt, February 2013. 671 Authors' Addresses 673 Per Hurtig 674 Karlstad University 675 Universitetsgatan 2 676 Karlstad 651 88 677 Sweden 679 Phone: +46 54 700 23 35 680 Email: per.hurtig@kau.se 682 Anna Brunstrom 683 Karlstad University 684 Universitetsgatan 2 685 Karlstad 651 88 686 Sweden 688 Phone: +46 54 700 17 95 689 Email: anna.brunstrom@kau.se 691 Andreas Petlund 692 Simula Research Laboratory AS 693 P.O. Box 134 694 Lysaker 1325 695 Norway 697 Phone: +47 67 82 82 00 698 Email: apetlund@simula.no 699 Michael Welzl 700 University of Oslo 701 PO Box 1080 Blindern 702 Oslo N-0316 703 Norway 705 Phone: +47 22 85 24 20 706 Email: michawe@ifi.uio.no