idnits 2.17.00 (12 Aug 2021) /tmp/idnits52205/draft-ietf-httpbis-priority-11.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 (9 December 2021) is 163 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: '-1' is mentioned on line 1291, but not defined -- Looks like a reference, but probably isn't: '6' on line 1291 -- Looks like a reference, but probably isn't: '0' on line 1291 -- Looks like a reference, but probably isn't: '7' on line 1291 -- Possible downref: Normative reference to a draft: ref. 'HTTP' == Outdated reference: A later version (-07) exists of draft-ietf-httpbis-http2bis-06 Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 HTTP K. Oku 3 Internet-Draft Fastly 4 Intended status: Standards Track L. Pardue 5 Expires: 12 June 2022 Cloudflare 6 9 December 2021 8 Extensible Prioritization Scheme for HTTP 9 draft-ietf-httpbis-priority-11 11 Abstract 13 This document describes a scheme that allows an HTTP client to 14 communicate its preferences for how the upstream server prioritizes 15 responses to its requests, and also allows a server to hint to a 16 downstream intermediary how its responses should be prioritized when 17 they are forwarded. This document defines the Priority header field 18 for communicating the initial priority in an HTTP version-independent 19 manner, as well as HTTP/2 and HTTP/3 frames for reprioritizing 20 responses. These share a common format structure that is designed to 21 provide future extensibility. 23 About This Document 25 This note is to be removed before publishing as an RFC. 27 Status information for this document may be found at 28 https://datatracker.ietf.org/doc/draft-ietf-httpbis-priority/. 30 Discussion of this document takes place on the HTTP Working Group 31 mailing list (mailto:ietf-http-wg@w3.org), which is archived at 32 https://lists.w3.org/Archives/Public/ietf-http-wg/. Working Group 33 information can be found at https://httpwg.org/. 35 Source for this draft and an issue tracker can be found at 36 https://github.com/httpwg/http-extensions/labels/priorities. 38 Status of This Memo 40 This Internet-Draft is submitted in full conformance with the 41 provisions of BCP 78 and BCP 79. 43 Internet-Drafts are working documents of the Internet Engineering 44 Task Force (IETF). Note that other groups may also distribute 45 working documents as Internet-Drafts. The list of current Internet- 46 Drafts is at https://datatracker.ietf.org/drafts/current/. 48 Internet-Drafts are draft documents valid for a maximum of six months 49 and may be updated, replaced, or obsoleted by other documents at any 50 time. It is inappropriate to use Internet-Drafts as reference 51 material or to cite them other than as "work in progress." 53 This Internet-Draft will expire on 12 June 2022. 55 Copyright Notice 57 Copyright (c) 2021 IETF Trust and the persons identified as the 58 document authors. All rights reserved. 60 This document is subject to BCP 78 and the IETF Trust's Legal 61 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 62 license-info) in effect on the date of publication of this document. 63 Please review these documents carefully, as they describe your rights 64 and restrictions with respect to this document. Code Components 65 extracted from this document must include Revised BSD License text as 66 described in Section 4.e of the Trust Legal Provisions and are 67 provided without warranty as described in the Revised BSD License. 69 Table of Contents 71 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 72 1.1. Notational Conventions . . . . . . . . . . . . . . . . . 5 73 2. Motivation for Replacing RFC 7540 Priorities . . . . . . . . 5 74 2.1. Disabling RFC 7540 Priorities . . . . . . . . . . . . . . 6 75 2.1.1. Advice when Using Extensible Priorities as the 76 Alternative . . . . . . . . . . . . . . . . . . . . . 7 77 3. Applicability of the Extensible Priority Scheme . . . . . . . 7 78 4. Priority Parameters . . . . . . . . . . . . . . . . . . . . . 8 79 4.1. Urgency . . . . . . . . . . . . . . . . . . . . . . . . . 8 80 4.2. Incremental . . . . . . . . . . . . . . . . . . . . . . . 9 81 4.3. Defining New Priority Parameters . . . . . . . . . . . . 10 82 4.3.1. Registration . . . . . . . . . . . . . . . . . . . . 10 83 5. The Priority HTTP Header Field . . . . . . . . . . . . . . . 11 84 6. Reprioritization . . . . . . . . . . . . . . . . . . . . . . 12 85 7. The PRIORITY_UPDATE Frame . . . . . . . . . . . . . . . . . . 12 86 7.1. HTTP/2 PRIORITY_UPDATE Frame . . . . . . . . . . . . . . 13 87 7.2. HTTP/3 PRIORITY_UPDATE Frame . . . . . . . . . . . . . . 14 88 8. Merging Client- and Server-Driven Priority Parameters . . . . 16 89 9. Client Scheduling . . . . . . . . . . . . . . . . . . . . . . 17 90 10. Server Scheduling . . . . . . . . . . . . . . . . . . . . . . 17 91 10.1. Intermediaries with Multiple Backend Connections . . . . 19 92 11. Scheduling and the CONNECT Method . . . . . . . . . . . . . . 19 93 12. Retransmission Scheduling . . . . . . . . . . . . . . . . . . 19 94 13. Fairness . . . . . . . . . . . . . . . . . . . . . . . . . . 20 95 13.1. Coalescing Intermediaries . . . . . . . . . . . . . . . 20 96 13.2. HTTP/1.x Back Ends . . . . . . . . . . . . . . . . . . . 21 97 13.3. Intentional Introduction of Unfairness . . . . . . . . . 21 98 14. Why use an End-to-End Header Field? . . . . . . . . . . . . . 21 99 15. Security Considerations . . . . . . . . . . . . . . . . . . . 22 100 16. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 22 101 17. References . . . . . . . . . . . . . . . . . . . . . . . . . 23 102 17.1. Normative References . . . . . . . . . . . . . . . . . . 23 103 17.2. Informative References . . . . . . . . . . . . . . . . . 24 104 Appendix A. Acknowledgements . . . . . . . . . . . . . . . . . . 25 105 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 26 106 B.1. Since draft-ietf-httpbis-priority-10 . . . . . . . . . . 26 107 B.2. Since draft-ietf-httpbis-priority-09 . . . . . . . . . . 26 108 B.3. Since draft-ietf-httpbis-priority-08 . . . . . . . . . . 26 109 B.4. Since draft-ietf-httpbis-priority-07 . . . . . . . . . . 26 110 B.5. Since draft-ietf-httpbis-priority-06 . . . . . . . . . . 26 111 B.6. Since draft-ietf-httpbis-priority-05 . . . . . . . . . . 27 112 B.7. Since draft-ietf-httpbis-priority-04 . . . . . . . . . . 27 113 B.8. Since draft-ietf-httpbis-priority-03 . . . . . . . . . . 27 114 B.9. Since draft-ietf-httpbis-priority-02 . . . . . . . . . . 27 115 B.10. Since draft-ietf-httpbis-priority-01 . . . . . . . . . . 27 116 B.11. Since draft-ietf-httpbis-priority-00 . . . . . . . . . . 28 117 B.12. Since draft-kazuho-httpbis-priority-04 . . . . . . . . . 28 118 B.13. Since draft-kazuho-httpbis-priority-03 . . . . . . . . . 28 119 B.14. Since draft-kazuho-httpbis-priority-02 . . . . . . . . . 28 120 B.15. Since draft-kazuho-httpbis-priority-01 . . . . . . . . . 29 121 B.16. Since draft-kazuho-httpbis-priority-00 . . . . . . . . . 29 122 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 29 124 1. Introduction 126 It is common for representations of an HTTP [HTTP] resource to have 127 relationships to one or more other resources. Clients will often 128 discover these relationships while processing a retrieved 129 representation, which may lead to further retrieval requests. 130 Meanwhile, the nature of the relationship determines whether the 131 client is blocked from continuing to process locally available 132 resources. An example of this is visual rendering of an HTML 133 document, which could be blocked by the retrieval of a CSS file that 134 the document refers to. In contrast, inline images do not block 135 rendering and get drawn incrementally as the chunks of the images 136 arrive. 138 HTTP/2 [HTTP2] and HTTP/3 [HTTP3] support multiplexing of requests 139 and responses in a single connection. An important feature of any 140 implementation of a protocol that provides multiplexing is the 141 ability to prioritize the sending of information. For example, to 142 provide meaningful presentation of an HTML document at the earliest 143 moment, it is important for an HTTP server to prioritize the HTTP 144 responses, or the chunks of those HTTP responses, that it sends to a 145 client. 147 A server that operates in ignorance of how clients issue requests and 148 consume responses can cause suboptimal client application 149 performance. Priority signals allow clients to communicate their 150 view of request priority. Servers have their own needs that are 151 independent from client needs, so they often combine priority signals 152 with other available information in order to inform scheduling of 153 response data. 155 RFC 7540 [RFC7540] stream priority allowed a client to send a series 156 of priority signals that communicate to the server a "priority tree"; 157 the structure of this tree represents the client's preferred relative 158 ordering and weighted distribution of the bandwidth among HTTP 159 responses. Servers could use these priority signals as input into 160 prioritization decision making. 162 The design and implementation of RFC 7540 stream priority was 163 observed to have shortcomings, explained in Section 2. HTTP/2 164 [HTTP2] has consequently deprecated the use of these stream priority 165 signals. The prioritization scheme and priority signals defined 166 herein can act as a substitute for RFC 7540 stream priority. 168 This document describes an extensible scheme for prioritizing HTTP 169 responses that uses absolute values. Section 4 defines priority 170 parameters, which are a standardized and extensible format of 171 priority information. Section 5 defines the Priority HTTP header 172 field, a protocol-version-independent and end-to-end priority signal. 173 Clients can send this header field to signal their view of how 174 responses should be prioritized. Similarly, servers behind an 175 intermediary can use it to signal priority to the intermediary. 176 After sending a request, a client can change the priority of the 177 response (see Section 6) using HTTP-version-specific frames defined 178 in Section 7.1 and Section 7.2. 180 Header field and frame priority signals are input to a server's 181 response prioritization process. They are only a suggestion and do 182 not guarantee any particular processing or transmission order for one 183 response relative to any other response. Section 10 and Section 12 184 provide consideration and guidance about how servers might act upon 185 signals. 187 1.1. Notational Conventions 189 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 190 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 191 "OPTIONAL" in this document are to be interpreted as described in 192 BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all 193 capitals, as shown here. 195 The terms Dictionary, sf-boolean, sf-dictionary, and sf-integer are 196 imported from [STRUCTURED-FIELDS]. 198 Example HTTP requests and responses use the HTTP/2-style formatting 199 from [HTTP2]. 201 This document uses the variable-length integer encoding from [QUIC]. 203 The term control stream is used to describe both the HTTP/2 stream 204 with identifier 0x0, and the HTTP/3 control stream; see Section 6.2.1 205 of [HTTP3]. 207 The term HTTP/2 priority signal is used to describe the priority 208 information sent from clients to servers in HTTP/2 frames; see 209 Section 5.3.2 of [HTTP2]. 211 2. Motivation for Replacing RFC 7540 Priorities 213 RFC 7540 stream priority (see Section 5.3 of [RFC7540]) is a complex 214 system where clients signal stream dependencies and weights to 215 describe an unbalanced tree. It suffered from limited deployment and 216 interoperability and was deprecated in a revision of HTTP/2 [HTTP2]. 217 HTTP/2 retains these protocol elements in order to maintain wire 218 compatibility (see Section 5.3.2 of [HTTP2]), which means that they 219 might still be used even in the presence of alternative signaling, 220 such as the scheme this document describes. 222 Many RFC 7540 server implementations do not act on HTTP/2 priority 223 signals. 225 Prioritization can use information that servers have about resources 226 or the order in which requests are generated. For example, a server, 227 with knowledge of an HTML document structure, might want to 228 prioritize the delivery of images that are critical to user 229 experience above other images. With RFC 7540 it is difficult for 230 servers to interpret signals from clients for prioritization as the 231 same conditions could result in very different signaling from 232 different clients. This document describes signaling that is simpler 233 and more constrained, requiring less interpretation and allowing less 234 variation. 236 RFC 7540 does not define a method that can be used by a server to 237 provide a priority signal for intermediaries. 239 RFC 7540 priority is expressed relative to other requests on the same 240 connection. Many requests are generated without knowledge of how 241 other requests might share a connection, which makes this difficult 242 to use reliably, especially in protocols that do not have strong 243 ordering guarantees, like HTTP/3 [HTTP3]. 245 Multiple experiments from independent research ([MARX], [MEENAN]) 246 have shown that simpler schemes can reach at least equivalent 247 performance characteristics compared to the more complex RFC 7540 248 setups seen in practice, at least for the web use case. 250 2.1. Disabling RFC 7540 Priorities 252 The problems and insights set out above provided the motivation for 253 an alternative to RFC 7540 stream priority (see Section 5.3 of 254 [HTTP2]). 256 The SETTINGS_NO_RFC7540_PRIORITIES HTTP/2 setting is defined by this 257 document in order to allow endpoints to omit or ignore HTTP/2 258 priority signals (see Section 5.3.2 of [HTTP2]), as described below. 259 The value of SETTINGS_NO_RFC7540_PRIORITIES MUST be 0 or 1. Any 260 value other than 0 or 1 MUST be treated as a connection error (see 261 Section 5.4.1 of [HTTP2]) of type PROTOCOL_ERROR. The initial value 262 is 0. 264 If endpoints use SETTINGS_NO_RFC7540_PRIORITIES they MUST send it in 265 the first SETTINGS frame. Senders MUST NOT change the 266 SETTINGS_NO_RFC7540_PRIORITIES value after the first SETTINGS frame. 267 Receivers that detect a change MAY treat it as a connection error of 268 type PROTOCOL_ERROR. 270 Clients can send SETTINGS_NO_RFC7540_PRIORITIES with a value of 1 to 271 indicate that they are not using HTTP/2 priority signals. The 272 SETTINGS frame precedes any HTTP/2 priority signal sent from clients, 273 so servers can determine whether they need to allocate any resources 274 to signal handling before signals arrive. A server that receives 275 SETTINGS_NO_RFC7540_PRIORITIES with a value of 1 MUST ignore HTTP/2 276 priority signals. 278 Servers can send SETTINGS_NO_RFC7540_PRIORITIES with a value of 1 to 279 indicate that they will ignore HTTP/2 priority signals sent by 280 clients. 282 Endpoints that send SETTINGS_NO_RFC7540_PRIORITIES are encouraged to 283 use alternative priority signals (for example, Section 5 or 284 Section 7.1) but there is no requirement to use a specific signal 285 type. 287 2.1.1. Advice when Using Extensible Priorities as the Alternative 289 Before receiving a SETTINGS frame from a server, a client does not 290 know if the server is ignoring HTTP/2 priority signals. Therefore, 291 until the client receives the SETTINGS frame from the server, the 292 client SHOULD send both the HTTP/2 priority signals and the signals 293 of this prioritization scheme (see Section 5 and Section 7.1). 295 Once the client receives the first SETTINGS frame that contains the 296 SETTINGS_NO_RFC7540_PRIORITIES parameter with value of 1, it SHOULD 297 stop sending the HTTP/2 priority signals. This avoids sending 298 redundant signals that are known to be ignored. 300 Similarly, if the client receives SETTINGS_NO_RFC7540_PRIORITIES with 301 value of 0 or if the settings parameter was absent, it SHOULD stop 302 sending PRIORITY_UPDATE frames (Section 7.1), since those frames are 303 likely to be ignored. However, the client MAY continue sending the 304 Priority header field (Section 5), as it is an end-to-end signal that 305 might be useful to nodes behind the server that the client is 306 directly connected to. 308 3. Applicability of the Extensible Priority Scheme 310 The priority scheme defined by this document is primarily focused on 311 the prioritization of HTTP response messages (see Section 3.4 of 312 [HTTP]). It defines new priority parameters (Section 4) and their 313 conveyors (Section 5 and Section 7) intended to communicate the 314 priority of responses to a server that is responsible for 315 prioritizing them. Section 10 provides considerations for servers 316 about acting on those signals in combination with other inputs and 317 factors. 319 The CONNECT method (see Section 9.3.6 of [HTTP]) can be used to 320 establish tunnels. Signaling applies similarly to tunnels; 321 additional considerations for server prioritization are given in 322 Section 11. 324 Section 9 describes how clients can optionally apply elements of this 325 scheme locally to the request messages that they generate. 327 Some forms of HTTP extensions might change HTTP/2 or HTTP/3 stream 328 behavior or define new data carriage mechanisms. Such extensions can 329 define themselves how this priority scheme is to be applied. 331 4. Priority Parameters 333 The priority information is a sequence of key-value pairs, providing 334 room for future extensions. Each key-value pair represents a 335 priority parameter. 337 The Priority HTTP header field (Section 5) is an end-to-end way to 338 transmit this set of priority parameters when a request or a response 339 is issued. In order to reprioritize a request (Section 6), HTTP- 340 version-specific PRIORITY_UPDATE frames (Section 7.1 and Section 7.2) 341 are used by clients to transmit the same information on a single hop. 343 Intermediaries can consume and produce priority signals in a 344 PRIORITY_UPDATE frame or Priority header field. Sending a 345 PRIORITY_UPDATE frame preserves the signal from the client, but 346 provides a signal that overrides that for the next hop; see 347 Section 14. Replacing or adding a Priority header field overrides 348 any signal from a client and can affect prioritization for all 349 subsequent recipients. 351 For both the Priority header field and the PRIORITY_UPDATE frame, the 352 set of priority parameters is encoded as a Structured Fields 353 Dictionary (see Section 3.2 of [STRUCTURED-FIELDS]). 355 This document defines the urgency(u) and incremental(i) priority 356 parameters. When receiving an HTTP request that does not carry these 357 priority parameters, a server SHOULD act as if their default values 358 were specified. 360 An intermediary can combine signals from requests and responses that 361 it forwards. Note that omission of priority parameters in responses 362 is handled differently from omission in requests; see Section 8. 364 Receivers parse the Dictionary as defined in Section 4.2 of 365 [STRUCTURED-FIELDS]. Where the Dictionary is successfully parsed, 366 this document places the additional requirement that unknown priority 367 parameters, priority parameters with out-of-range values, or values 368 of unexpected types MUST be ignored. 370 4.1. Urgency 372 The urgency parameter (u) takes an integer between 0 and 7, in 373 descending order of priority. 375 The value is encoded as an sf-integer. The default value is 3. 377 Endpoints use this parameter to communicate their view of the 378 precedence of HTTP responses. The chosen value of urgency can be 379 based on the expectation that servers might use this information to 380 transmit HTTP responses in the order of their urgency. The smaller 381 the value, the higher the precedence. 383 The following example shows a request for a CSS file with the urgency 384 set to 0: 386 :method = GET 387 :scheme = https 388 :authority = example.net 389 :path = /style.css 390 priority = u=0 392 A client that fetches a document that likely consists of multiple 393 HTTP resources (e.g., HTML) SHOULD assign the default urgency level 394 to the main resource. This convention allows servers to refine the 395 urgency using knowledge specific to the web-site (see Section 8). 397 The lowest urgency level (7) is reserved for background tasks such as 398 delivery of software updates. This urgency level SHOULD NOT be used 399 for fetching responses that have impact on user interaction. 401 4.2. Incremental 403 The incremental parameter (i) takes an sf-boolean as the value that 404 indicates if an HTTP response can be processed incrementally, i.e., 405 provide some meaningful output as chunks of the response arrive. 407 The default value of the incremental parameter is false (0). 409 If a client makes concurrent requests with the incremental parameter 410 set to false, there is no benefit serving responses with the same 411 urgency concurrently because the client is not going to process those 412 responses incrementally. Serving non-incremental responses with the 413 same urgency one by one, in the order in which those requests were 414 generated is considered to be the best strategy. 416 If a client makes concurrent requests with the incremental parameter 417 set to true, serving requests with the same urgency concurrently 418 might be beneficial. Doing this distributes the connection 419 bandwidth, meaning that responses take longer to complete. 420 Incremental delivery is most useful where multiple partial responses 421 might provide some value to clients ahead of a complete response 422 being available. 424 The following example shows a request for a JPEG file with the 425 urgency parameter set to 5 and the incremental parameter set to true. 427 :method = GET 428 :scheme = https 429 :authority = example.net 430 :path = /image.jpg 431 priority = u=5, i 433 4.3. Defining New Priority Parameters 435 When attempting to define new priority parameters, care must be taken 436 so that they do not adversely interfere with prioritization performed 437 by existing endpoints or intermediaries that do not understand the 438 newly defined priority parameter. Since unknown priority parameters 439 are ignored, new priority parameters should not change the 440 interpretation of, or modify, the urgency (see Section 4.1) or 441 incremental (see Section 4.2) priority parameters in a way that is 442 not backwards compatible or fallback safe. 444 For example, if there is a need to provide more granularity than 445 eight urgency levels, it would be possible to subdivide the range 446 using an additional priority parameter. Implementations that do not 447 recognize the parameter can safely continue to use the less granular 448 eight levels. 450 Alternatively, the urgency can be augmented. For example, a 451 graphical user agent could send a visible priority parameter to 452 indicate if the resource being requested is within the viewport. 454 Generic priority parameters are preferred over vendor-specific, 455 application-specific or deployment-specific values. If a generic 456 value cannot be agreed upon in the community, the parameter's name 457 should be correspondingly specific (e.g., with a prefix that 458 identifies the vendor, application or deployment). 460 4.3.1. Registration 462 New priority parameters can be defined by registering them in the 463 HTTP Priority Parameters Registry. The registry governs the keys 464 (short textual strings) used in the Structured Fields Dictionary (see 465 Section 3.2 of [STRUCTURED-FIELDS]). Since each HTTP request can 466 have associated priority signals, there is value in having short key 467 lengths, especially single-character strings. In order to encourage 468 extension while avoiding unintended conflict among attractive key 469 values, the HTTP Priority Parameters Registry operates two 470 registration policies depending on key length. 472 * Registration requests for priority parameters with a key length of 473 one use the Specification Required policy, as per Section 4.6 of 474 [RFC8126]. 476 * Registration requests for priority parameters with a key length 477 greater than one use the Expert Review policy, as per Section 4.5 478 of [RFC8126]. A specification document is appreciated, but not 479 required. 481 When reviewing registration requests, the designated expert(s) can 482 consider the additional guidance provided in Section 4.3 but cannot 483 use it as a basis for rejection. 485 Registration requests should use the following template: 487 Name: [a name for the Priority Parameter that matches key] 489 Description: [a description of the priority parameter semantics and 490 value] 492 Reference: [to a specification defining this priority parameter] 494 See the registry at https://iana.org/assignments/http-priority 495 (https://iana.org/assignments/http-priority) for details on where to 496 send registration requests. 498 5. The Priority HTTP Header Field 500 The Priority HTTP header field carries priority parameters (see 501 Section 4). It can appear in requests and responses. It is an end- 502 to-end signal of the request priority from the client or the response 503 priority from the server. Section 8 describes how intermediaries can 504 combine the priority information sent from clients and servers. 505 Clients cannot interpret the appearance or omission of a Priority 506 response header field as acknowledgement that any prioritization has 507 occurred. Guidance for how endpoints can act on Priority header 508 values is given in Section 10 and Section 9. 510 Priority is a Dictionary (Section 3.2 of [STRUCTURED-FIELDS]): 512 Priority = sf-dictionary 513 An HTTP request with a Priority header field might be cached and re- 514 used for subsequent requests; see [CACHING]. When an origin server 515 generates the Priority response header field based on properties of 516 an HTTP request it receives, the server is expected to control the 517 cacheability or the applicability of the cached response, by using 518 header fields that control the caching behavior (e.g., Cache-Control, 519 Vary). 521 6. Reprioritization 523 After a client sends a request, it may be beneficial to change the 524 priority of the response. As an example, a web browser might issue a 525 prefetch request for a JavaScript file with the urgency parameter of 526 the Priority request header field set to u=7 (background). Then, 527 when the user navigates to a page which references the new JavaScript 528 file, while the prefetch is in progress, the browser would send a 529 reprioritization signal with the priority field value set to u=0. 530 The PRIORITY_UPDATE frame (Section 7) can be used for such 531 reprioritization. 533 7. The PRIORITY_UPDATE Frame 535 This document specifies a new PRIORITY_UPDATE frame for HTTP/2 536 [HTTP2] and HTTP/3 [HTTP3]. It carries priority parameters and 537 references the target of the prioritization based on a version- 538 specific identifier. In HTTP/2, this identifier is the Stream ID; in 539 HTTP/3, the identifier is either the Stream ID or Push ID. Unlike 540 the Priority header field, the PRIORITY_UPDATE frame is a hop-by-hop 541 signal. 543 PRIORITY_UPDATE frames are sent by clients on the control stream, 544 allowing them to be sent independent from the stream that carries the 545 response. This means they can be used to reprioritize a response or 546 a push stream; or signal the initial priority of a response instead 547 of the Priority header field. 549 A PRIORITY_UPDATE frame communicates a complete set of all priority 550 parameters in the Priority Field Value field. Omitting a priority 551 parameter is a signal to use its default value. Failure to parse the 552 Priority Field Value MAY be treated as a connection error. In HTTP/2 553 the error is of type PROTOCOL_ERROR; in HTTP/3 the error is of type 554 H3_GENERAL_PROTOCOL_ERROR. 556 A client MAY send a PRIORITY_UPDATE frame before the stream that it 557 references is open (except for HTTP/2 push streams; see Section 7.1). 558 Furthermore, HTTP/3 offers no guaranteed ordering across streams, 559 which could cause the frame to be received earlier than intended. 560 Either case leads to a race condition where a server receives a 561 PRIORITY_UPDATE frame that references a request stream that is yet to 562 be opened. To solve this condition, for the purposes of scheduling, 563 the most recently received PRIORITY_UPDATE frame can be considered as 564 the most up-to-date information that overrides any other signal. 565 Servers SHOULD buffer the most recently received PRIORITY_UPDATE 566 frame and apply it once the referenced stream is opened. Holding 567 PRIORITY_UPDATE frames for each stream requires server resources, 568 which can can be bounded by local implementation policy. Although 569 there is no limit to the number of PRIORITY_UPDATES that can be sent, 570 storing only the most recently received frame limits resource 571 commitment. 573 7.1. HTTP/2 PRIORITY_UPDATE Frame 575 The HTTP/2 PRIORITY_UPDATE frame (type=0x10) is used by clients to 576 signal the initial priority of a response, or to reprioritize a 577 response or push stream. It carries the stream ID of the response 578 and the priority in ASCII text, using the same representation as the 579 Priority header field value. 581 The Stream Identifier field (see Section 5.1.1 of [HTTP2]) in the 582 PRIORITY_UPDATE frame header MUST be zero (0x0). Receiving a 583 PRIORITY_UPDATE frame with a field of any other value MUST be treated 584 as a connection error of type PROTOCOL_ERROR. 586 HTTP/2 PRIORITY_UPDATE Frame { 587 Length (24), 588 Type (i) = 10, 590 Unused Flags (8). 592 Reserved (1), 593 Stream Identifier (31), 595 Reserved (1), 596 Prioritized Stream ID (31), 597 Priority Field Value (..), 598 } 600 Figure 1: HTTP/2 PRIORITY_UPDATE Frame Payload 602 The Length, Type, Unused Flag(s), Reserved, and Stream Identifier 603 fields are described in Section 4 of [HTTP2]. The frame payload of 604 PRIORITY_UPDATE frame payload contains the following additional 605 fields: 607 Reserved: A reserved 1-bit field. The semantics of this bit are 608 undefined, and the bit MUST remain unset (0x0) when sending and 609 MUST be ignored when receiving. 611 Prioritized Stream ID: A 31-bit stream identifier for the stream 612 that is the target of the priority update. 614 Priority Field Value: The priority update value in ASCII text, 615 encoded using Structured Fields. This is the same representation 616 as the Priority header field value. 618 When the PRIORITY_UPDATE frame applies to a request stream, clients 619 SHOULD provide a Prioritized Stream ID that refers to a stream in the 620 "open", "half-closed (local)", or "idle" state. Servers can discard 621 frames where the Prioritized Stream ID refers to a stream in the 622 "half-closed (local)" or "closed" state. The number of streams which 623 have been prioritized but remain in the "idle" state plus the number 624 of active streams (those in the "open" or either "half-closed" state; 625 see Section 5.1.2 of [HTTP2]) MUST NOT exceed the value of the 626 SETTINGS_MAX_CONCURRENT_STREAMS parameter. Servers that receive such 627 a PRIORITY_UPDATE MUST respond with a connection error of type 628 PROTOCOL_ERROR. 630 When the PRIORITY_UPDATE frame applies to a push stream, clients 631 SHOULD provide a Prioritized Stream ID that refers to a stream in the 632 "reserved (remote)" or "half-closed (local)" state. Servers can 633 discard frames where the Prioritized Stream ID refers to a stream in 634 the "closed" state. Clients MUST NOT provide a Prioritized Stream ID 635 that refers to a push stream in the "idle" state. Servers that 636 receive a PRIORITY_UPDATE for a push stream in the "idle" state MUST 637 respond with a connection error of type PROTOCOL_ERROR. 639 If a PRIORITY_UPDATE frame is received with a Prioritized Stream ID 640 of 0x0, the recipient MUST respond with a connection error of type 641 PROTOCOL_ERROR. 643 If a client receives a PRIORITY_UPDATE frame, it MUST respond with a 644 connection error of type PROTOCOL_ERROR. 646 7.2. HTTP/3 PRIORITY_UPDATE Frame 648 The HTTP/3 PRIORITY_UPDATE frame (type=0xF0700 or 0xF0701) is used by 649 clients to signal the initial priority of a response, or to 650 reprioritize a response or push stream. It carries the identifier of 651 the element that is being prioritized, and the updated priority in 652 ASCII text, using the same representation as that of the Priority 653 header field value. PRIORITY_UPDATE with a frame type of 0xF0700 is 654 used for request streams, while PRIORITY_UPDATE with a frame type of 655 0xF0701 is used for push streams. 657 The PRIORITY_UPDATE frame MUST be sent on the client control stream 658 (see Section 6.2.1 of [HTTP3]). Receiving a PRIORITY_UPDATE frame on 659 a stream other than the client control stream MUST be treated as a 660 connection error of type H3_FRAME_UNEXPECTED. 662 HTTP/3 PRIORITY_UPDATE Frame { 663 Type (i) = 0xF0700..0xF0701, 664 Length (i), 665 Prioritized Element ID (i), 666 Priority Field Value (..), 667 } 669 Figure 2: HTTP/3 PRIORITY_UPDATE Frame 671 The PRIORITY_UPDATE frame payload has the following fields: 673 Prioritized Element ID: The stream ID or push ID that is the target 674 of the priority update. 676 Priority Field Value: The priority update value in ASCII text, 677 encoded using Structured Fields. This is the same representation 678 as the Priority header field value. 680 The request-stream variant of PRIORITY_UPDATE (type=0xF0700) MUST 681 reference a request stream. If a server receives a PRIORITY_UPDATE 682 (type=0xF0700) for a Stream ID that is not a request stream, this 683 MUST be treated as a connection error of type H3_ID_ERROR. The 684 Stream ID MUST be within the client-initiated bidirectional stream 685 limit. If a server receives a PRIORITY_UPDATE (type=0xF0700) with a 686 Stream ID that is beyond the stream limits, this SHOULD be treated as 687 a connection error of type H3_ID_ERROR. Generating an error is not 688 mandatory because HTTP/3 implementations might have practical 689 barriers to determining the active stream concurrency limit that is 690 applied by the QUIC layer. 692 The push-stream variant PRIORITY_UPDATE (type=0xF0701) MUST reference 693 a promised push stream. If a server receives a PRIORITY_UPDATE 694 (type=0xF0701) with a Push ID that is greater than the maximum Push 695 ID or which has not yet been promised, this MUST be treated as a 696 connection error of type H3_ID_ERROR. 698 PRIORITY_UPDATE frames of either type are only sent by clients. If a 699 client receives a PRIORITY_UPDATE frame, this MUST be treated as a 700 connection error of type H3_FRAME_UNEXPECTED. 702 8. Merging Client- and Server-Driven Priority Parameters 704 It is not always the case that the client has the best understanding 705 of how the HTTP responses deserve to be prioritized. The server 706 might have additional information that can be combined with the 707 client's indicated priority in order to improve the prioritization of 708 the response. For example, use of an HTML document might depend 709 heavily on one of the inline images; existence of such dependencies 710 is typically best known to the server. Or, a server that receives 711 requests for a font [RFC8081] and images with the same urgency might 712 give higher precedence to the font, so that a visual client can 713 render textual information at an early moment. 715 An origin can use the Priority response header field to indicate its 716 view on how an HTTP response should be prioritized. An intermediary 717 that forwards an HTTP response can use the priority parameters found 718 in the Priority response header field, in combination with the client 719 Priority request header field, as input to its prioritization 720 process. No guidance is provided for merging priorities, this is 721 left as an implementation decision. 723 Absence of a priority parameter in an HTTP response indicates the 724 server's disinterest in changing the client-provided value. This is 725 different from the the request header field, in which omission of a 726 priority parameter implies the use of their default values (see 727 Section 4). 729 As a non-normative example, when the client sends an HTTP request 730 with the urgency parameter set to 5 and the incremental parameter set 731 to true 733 :method = GET 734 :scheme = https 735 :authority = example.net 736 :path = /menu.png 737 priority = u=5, i 739 and the origin responds with 741 :status = 200 742 content-type = image/png 743 priority = u=1 745 the intermediary might alter its understanding of the urgency from 5 746 to 1, because it prefers the server-provided value over the client's. 747 The incremental value continues to be true, the value specified by 748 the client, as the server did not specify the incremental(i) 749 parameter. 751 9. Client Scheduling 753 A client MAY use priority values to make local processing or 754 scheduling choices about the requests it initiates. 756 10. Server Scheduling 758 It is generally beneficial for an HTTP server to send all responses 759 as early as possible. However, when serving multiple requests on a 760 single connection, there could be competition between the requests 761 for resources such as connection bandwidth. This section describes 762 considerations regarding how servers can schedule the order in which 763 the competing responses will be sent, when such competition exists. 765 Server scheduling is a prioritization process based on many inputs, 766 with priority signals being only one form of input. Factors such as 767 implementation choices or deployment environment also play a role. 768 Any given connection is likely to have many dynamic permutations. 769 For these reasons, there is no unilateral perfect scheduler. This 770 document provides some basic, non-exhaustive, recommendations for how 771 servers might act on priority parameters. It does not describe in 772 detail how servers might combine priority signals with other factors. 773 Endpoints cannot depend on particular treatment based on priority 774 signals. Expressing priority is only a suggestion. 776 It is RECOMMENDED that, when possible, servers respect the urgency 777 parameter (Section 4.1), sending higher urgency responses before 778 lower urgency responses. 780 The incremental parameter indicates how a client processes response 781 bytes as they arrive. It is RECOMMENDED that, when possible, servers 782 respect the incremental parameter (Section 4.2). 784 Non-incremental responses of the same urgency SHOULD be served by 785 prioritizing bandwidth allocation in ascending order of the stream 786 ID, which corresponds to the order in which clients make requests. 787 Doing so ensures that clients can use request ordering to influence 788 response order. 790 Incremental responses of the same urgency SHOULD be served by sharing 791 bandwidth amongst them. Incremental resources are used as parts, or 792 chunks, of the response payload are received. A client might benefit 793 more from receiving a portion of all these resources rather than the 794 entirety of a single resource. How large a portion of the resource 795 is needed to be useful in improving performance varies. Some 796 resource types place critical elements early, others can use 797 information progressively. This scheme provides no explicit mandate 798 about how a server should use size, type or any other input to decide 799 how to prioritize. 801 There can be scenarios where a server will need to schedule multiple 802 incremental and non-incremental responses at the same urgency level. 803 Strictly abiding the scheduling guidance based on urgency and request 804 generation order might lead to sub-optimal results at the client, as 805 early non-incremental responses might prevent serving of incremental 806 responses issued later. The following are examples of such 807 challenges. 809 1. At the same urgency level, a non-incremental request for a large 810 resource followed by an incremental request for a small resource. 812 2. At the same urgency level, an incremental request of 813 indeterminate length followed by a non-incremental large 814 resource. 816 It is RECOMMENDED that servers avoid such starvation where possible. 817 The method to do so is an implementation decision. For example, a 818 server might pre-emptively send responses of a particular incremental 819 type based on other information such as content size. 821 Optimal scheduling of server push is difficult, especially when 822 pushed resources contend with active concurrent requests. Servers 823 can consider many factors when scheduling, such as the type or size 824 of resource being pushed, the priority of the request that triggered 825 the push, the count of active concurrent responses, the priority of 826 other active concurrent responses, etc. There is no general guidance 827 on the best way to apply these. A server that is too simple could 828 easily push at too high a priority and block client requests, or push 829 at too low a priority and delay the response, negating intended goals 830 of server push. 832 Priority signals are a factor for server push scheduling. The 833 concept of parameter value defaults applies slightly differently 834 because there is no explicit client-signalled initial priority. A 835 server can apply priority signals provided in an origin response; see 836 the merging guidance given in Section 8. In the absence of origin 837 signals, applying default parameter values could be suboptimal. By 838 whatever means a server decides to schedule a pushed response, it can 839 signal the intended priority to the client by including the Priority 840 field in a PUSH_PROMISE or HEADERS frame. 842 10.1. Intermediaries with Multiple Backend Connections 844 An intermediary serving an HTTP connection might split requests over 845 multiple backend connections. When it applies prioritization rules 846 strictly, low priority requests cannot make progress while requests 847 with higher priorities are in flight. This blocking can propagate to 848 backend connections, which the peer might interpret as a connection 849 stall. Endpoints often implement protections against stalls, such as 850 abruptly closing connections after a certain time period. To reduce 851 the possibility of this occurring, intermediaries can avoid strictly 852 following prioritization and instead allocate small amounts of 853 bandwidth for all the requests that they are forwarding, so that 854 every request can make some progress over time. 856 Similarly, servers SHOULD allocate some amount of bandwidths to 857 streams acting as tunnels. 859 11. Scheduling and the CONNECT Method 861 When a request stream carries the CONNECT method, the scheduling 862 guidance in this document applies to the frames on the stream. A 863 client that issues multiple CONNECT requests can set the incremental 864 parameter to true. Servers that implement the recommendations for 865 handling of the incremental parameter in Section 10 are likely to 866 schedule these fairly, avoiding one CONNECT stream from blocking 867 others. 869 12. Retransmission Scheduling 871 Transport protocols such as TCP and QUIC provide reliability by 872 detecting packet losses and retransmitting lost information. In 873 addition to the considerations in Section 10, scheduling of 874 retransmission data could compete with new data. The remainder of 875 this section discusses considerations when using QUIC. 877 Section 13.3 of [QUIC] states "Endpoints SHOULD prioritize 878 retransmission of data over sending new data, unless priorities 879 specified by the application indicate otherwise". When an HTTP/3 880 application uses the priority scheme defined in this document and the 881 QUIC transport implementation supports application indicated stream 882 priority, a transport that considers the relative priority of streams 883 when scheduling both new data and retransmission data might better 884 match the expectations of the application. However, there are no 885 requirements on how a transport chooses to schedule based on this 886 information because the decision depends on several factors and 887 trade-offs. It could prioritize new data for a higher urgency stream 888 over retransmission data for a lower priority stream, or it could 889 prioritize retransmission data over new data irrespective of 890 urgencies. 892 Section 6.2.4 of [QUIC-RECOVERY] also highlights consideration of 893 application priorities when sending probe packets after Probe Timeout 894 timer expiration. A QUIC implementation supporting application- 895 indicated priorities might use the relative priority of streams when 896 choosing probe data. 898 13. Fairness 900 Typically, HTTP implementations depend on the underlying transport to 901 maintain fairness between connections competing for bandwidth. When 902 HTTP requests are forwarded through intermediaries, progress made by 903 each connection originating from end clients can become different 904 over time, depending on how intermediaries coalesce or split requests 905 into backend connections. This unfairness can expand if priority 906 signals are used. Section 13.1 and Section 13.2 discuss mitigations 907 against this expansion of unfairness. 909 Conversely, Section 13.3 discusses how servers might intentionally 910 allocate unequal bandwidth to some connections depending on the 911 priority signals. 913 13.1. Coalescing Intermediaries 915 When an intermediary coalesces HTTP requests coming from multiple 916 clients into one HTTP/2 or HTTP/3 connection going to the backend 917 server, requests that originate from one client might carry signals 918 indicating higher priority than those coming from others. 920 It is sometimes beneficial for the server running behind an 921 intermediary to obey Priority header field values. As an example, a 922 resource-constrained server might defer the transmission of software 923 update files that would have the background urgency being associated. 924 However, in the worst case, the asymmetry between the priority 925 declared by multiple clients might cause responses going to one user 926 agent to be delayed totally after those going to another. 928 In order to mitigate this fairness problem, a server could use 929 knowledge about the intermediary as another input in its 930 prioritization decisions. For instance, if a server knows the 931 intermediary is coalescing requests, then it could avoid serving the 932 responses in their entirety and instead distribute bandwidth (for 933 example, in a round-robin manner). This can work if the constrained 934 resource is network capacity between the intermediary and the user 935 agent, as the intermediary buffers responses and forwards the chunks 936 based on the prioritization scheme it implements. 938 A server can determine if a request came from an intermediary through 939 configuration, or by consulting if that request contains one of the 940 following header fields: 942 * Forwarded [FORWARDED], X-Forwarded-For 944 * Via (see Section 7.6.3 of [HTTP]) 946 13.2. HTTP/1.x Back Ends 948 It is common for CDN infrastructure to support different HTTP 949 versions on the front end and back end. For instance, the client- 950 facing edge might support HTTP/2 and HTTP/3 while communication to 951 back end servers is done using HTTP/1.1. Unlike with connection 952 coalescing, the CDN will "de-mux" requests into discrete connections 953 to the back end. HTTP/1.1 and older do not support response 954 multiplexing in a single connection, so there is not a fairness 955 problem. However, back end servers MAY still use client headers for 956 request scheduling. Back end servers SHOULD only schedule based on 957 client priority information where that information can be scoped to 958 individual end clients. Authentication and other session information 959 might provide this linkability. 961 13.3. Intentional Introduction of Unfairness 963 It is sometimes beneficial to deprioritize the transmission of one 964 connection over others, knowing that doing so introduces a certain 965 amount of unfairness between the connections and therefore between 966 the requests served on those connections. 968 For example, a server might use a scavenging congestion controller on 969 connections that only convey background priority responses such as 970 software update images. Doing so improves responsiveness of other 971 connections at the cost of delaying the delivery of updates. 973 14. Why use an End-to-End Header Field? 975 In contrast to the prioritization scheme of HTTP/2 that uses a hop- 976 by-hop frame, the Priority header field is defined as end-to-end. 978 The way that a client processes a response is a property associated 979 with the client generating that request. Not that of an 980 intermediary. Therefore, it is an end-to-end property. How these 981 end-to-end properties carried by the Priority header field affect the 982 prioritization between the responses that share a connection is a 983 hop-by-hop issue. 985 Having the Priority header field defined as end-to-end is important 986 for caching intermediaries. Such intermediaries can cache the value 987 of the Priority header field along with the response, and utilize the 988 value of the cached header field when serving the cached response, 989 only because the header field is defined as end-to-end rather than 990 hop-by-hop. 992 15. Security Considerations 994 Section 7 describes considerations for server buffering of 995 PRIORITY_UPDATE frames. 997 Section 10 presents examples where servers that prioritize responses 998 in a certain way might be starved of the ability to transmit payload. 1000 The security considerations from [STRUCTURED-FIELDS] apply to 1001 processing of priority parameters defined in Section 4. 1003 16. IANA Considerations 1005 This specification registers the following entry in the the Hypertext 1006 Transfer Protocol (HTTP) Field Name Registry established by [HTTP]: 1008 Field name: Priority 1010 Status: permanent 1012 Specification document(s): This document 1014 This specification registers the following entry in the HTTP/2 1015 Settings registry established by [RFC7540]: 1017 Name: SETTINGS_NO_RFC7540_PRIORITIES 1019 Code: 0x9 1021 Initial value: 0 1023 Specification: This document 1024 This specification registers the following entry in the HTTP/2 Frame 1025 Type registry established by [RFC7540]: 1027 Frame Type: PRIORITY_UPDATE 1029 Code: 0x10 1031 Specification: This document 1033 This specification registers the following entries in the HTTP/3 1034 Frame Type registry established by [HTTP3]: 1036 Frame Type: PRIORITY_UPDATE 1038 Code: 0xF0700 and 0xF0701 1040 Specification: This document 1042 Upon publication, please create the HTTP Priority Parameters registry 1043 at https://iana.org/assignments/http-priority 1044 (https://iana.org/assignments/http-priority) and populate it with the 1045 entries in Table 1; see Section 4.3.1 for its associated procedures. 1047 +======+==================================+===============+ 1048 | Name | Description | Specification | 1049 +======+==================================+===============+ 1050 | u | The urgency of an HTTP response. | Section 4.1 | 1051 +------+----------------------------------+---------------+ 1052 | i | Whether an HTTP response can be | Section 4.2 | 1053 | | processed incrementally. | | 1054 +------+----------------------------------+---------------+ 1056 Table 1: Initial Priority Parameters 1058 17. References 1060 17.1. Normative References 1062 [HTTP] Fielding, R. T., Nottingham, M., and J. Reschke, "HTTP 1063 Semantics", Work in Progress, Internet-Draft, draft-ietf- 1064 httpbis-semantics-19, 12 September 2021, 1065 . 1068 [HTTP2] Thomson, M. and C. Benfield, "Hypertext Transfer Protocol 1069 Version 2 (HTTP/2)", Work in Progress, Internet-Draft, 1070 draft-ietf-httpbis-http2bis-06, 18 November 2021, 1071 . 1074 [HTTP3] Bishop, M., "Hypertext Transfer Protocol Version 3 1075 (HTTP/3)", Work in Progress, Internet-Draft, draft-ietf- 1076 quic-http-34, 2 February 2021, 1077 . 1080 [QUIC] Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based 1081 Multiplexed and Secure Transport", RFC 9000, 1082 DOI 10.17487/RFC9000, May 2021, 1083 . 1085 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1086 Requirement Levels", BCP 14, RFC 2119, 1087 DOI 10.17487/RFC2119, March 1997, 1088 . 1090 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 1091 Writing an IANA Considerations Section in RFCs", BCP 26, 1092 RFC 8126, DOI 10.17487/RFC8126, June 2017, 1093 . 1095 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 1096 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 1097 May 2017, . 1099 [STRUCTURED-FIELDS] 1100 Nottingham, M. and P-H. Kamp, "Structured Field Values for 1101 HTTP", RFC 8941, DOI 10.17487/RFC8941, February 2021, 1102 . 1104 17.2. Informative References 1106 [CACHING] Fielding, R. T., Nottingham, M., and J. Reschke, "HTTP 1107 Caching", Work in Progress, Internet-Draft, draft-ietf- 1108 httpbis-cache-19, 12 September 2021, 1109 . 1112 [FORWARDED] 1113 Petersson, A. and M. Nilsson, "Forwarded HTTP Extension", 1114 RFC 7239, DOI 10.17487/RFC7239, June 2014, 1115 . 1117 [I-D.lassey-priority-setting] 1118 Lassey, B. and L. Pardue, "Declaring Support for HTTP/2 1119 Priorities", Work in Progress, Internet-Draft, draft- 1120 lassey-priority-setting-00, 25 July 2019, 1121 . 1124 [MARX] Marx, R., Decker, T.D., Quax, P., and W. Lamotte, "Of the 1125 Utmost Importance: Resource Prioritization in HTTP/3 over 1126 QUIC", DOI 10.5220/0008191701300143, 1127 SCITEPRESS Proceedings of the 15th International 1128 Conference on Web Information Systems and Technologies 1129 (pages 130-143), September 2019, 1130 . 1132 [MEENAN] Meenan, P., "Better HTTP/2 Prioritization for a Faster 1133 Web", 14 May 2019, . 1136 [QUIC-RECOVERY] 1137 Iyengar, J., Ed. and I. Swett, Ed., "QUIC Loss Detection 1138 and Congestion Control", RFC 9002, DOI 10.17487/RFC9002, 1139 May 2021, . 1141 [RFC7540] Belshe, M., Peon, R., and M. Thomson, Ed., "Hypertext 1142 Transfer Protocol Version 2 (HTTP/2)", RFC 7540, 1143 DOI 10.17487/RFC7540, May 2015, 1144 . 1146 [RFC8081] Lilley, C., "The "font" Top-Level Media Type", RFC 8081, 1147 DOI 10.17487/RFC8081, February 2017, 1148 . 1150 Appendix A. Acknowledgements 1152 Roy Fielding presented the idea of using a header field for 1153 representing priorities in http://tools.ietf.org/agenda/83/slides/ 1154 slides-83-httpbis-5.pdf (http://tools.ietf.org/agenda/83/slides/ 1155 slides-83-httpbis-5.pdf). In https://github.com/pmeenan/http3- 1156 prioritization-proposal (https://github.com/pmeenan/http3- 1157 prioritization-proposal), Patrick Meenan advocated for representing 1158 the priorities using a tuple of urgency and concurrency. The ability 1159 to disable HTTP/2 prioritization is inspired by 1160 [I-D.lassey-priority-setting], authored by Brad Lassey and Lucas 1161 Pardue, with modifications based on feedback that was not 1162 incorporated into an update to that document. 1164 The motivation for defining an alternative to HTTP/2 priorities is 1165 drawn from discussion within the broad HTTP community. Special 1166 thanks to Roberto Peon, Martin Thomson and Netflix for text that was 1167 incorporated explicitly in this document. 1169 In addition to the people above, this document owes a lot to the 1170 extensive discussion in the HTTP priority design team, consisting of 1171 Alan Frindell, Andrew Galloni, Craig Taylor, Ian Swett, Kazuho Oku, 1172 Lucas Pardue, Matthew Cox, Mike Bishop, Roberto Peon, Robin Marx, Roy 1173 Fielding. 1175 Yang Chi contributed the section on retransmission scheduling. 1177 Appendix B. Change Log 1179 _RFC EDITOR: please remove this section before publication_ 1181 B.1. Since draft-ietf-httpbis-priority-10 1183 * Editorial changes 1185 * Add clearer IANA instructions for Priority Parameter initial 1186 population 1188 B.2. Since draft-ietf-httpbis-priority-09 1190 * Editorial changes 1192 B.3. Since draft-ietf-httpbis-priority-08 1194 * Changelog fixups 1196 B.4. Since draft-ietf-httpbis-priority-07 1198 * Relax requirements of receiving SETTINGS_NO_RFC7540_PRIORITIES 1199 that changes value (#1714, #1725) 1201 * Clarify how intermediaries might use frames vs. headers (#1715, 1202 #1735) 1204 * Relax requirement when receiving a PRIORITY_UPDATE with an invalid 1205 structured field value (#1741, #1756) 1207 B.5. Since draft-ietf-httpbis-priority-06 1209 * Focus on editorial changes 1211 * Clarify rules about Sf-Dictionary handling in headers 1212 * Split policy for parameter IANA registry into two sections based 1213 on key length 1215 B.6. Since draft-ietf-httpbis-priority-05 1217 * Renamed SETTINGS_DEPRECATE_RFC7540_PRIORITIES to 1218 SETTINGS_NO_RFC7540_PRIORITIES 1220 * Clarify that senders of the HTTP/2 setting can use any alternative 1221 (#1679, #1705) 1223 B.7. Since draft-ietf-httpbis-priority-04 1225 * Renamed SETTINGS_DEPRECATE_HTTP2_PRIORITIES to 1226 SETTINGS_DEPRECATE_RFC7540_PRIORITIES (#1601) 1228 * Reoriented text towards RFC7540bis (#1561, #1601) 1230 * Clarify intermediary behavior (#1562) 1232 B.8. Since draft-ietf-httpbis-priority-03 1234 * Add statement about what this scheme applies to. Clarify 1235 extensions can use it but must define how themselves (#1550, 1236 #1559) 1238 * Describe scheduling considerations for the CONNECT method (#1495, 1239 #1544) 1241 * Describe scheduling considerations for retransmitted data (#1429, 1242 #1504) 1244 * Suggest intermediaries might avoid strict prioritization (#1562) 1246 B.9. Since draft-ietf-httpbis-priority-02 1248 * Describe considerations for server push prioritization (#1056, 1249 #1345) 1251 * Define HTTP/2 PRIORITY_UPDATE ID limits in HTTP/2 terms (#1261, 1252 #1344) 1254 * Add a Priority Parameters registry (#1371) 1256 B.10. Since draft-ietf-httpbis-priority-01 1258 * PRIORITY_UPDATE frame changes (#1096, #1079, #1167, #1262, #1267, 1259 #1271) 1261 * Add section to describe server scheduling considerations (#1215, 1262 #1232, #1266) 1264 * Remove specific instructions related to intermediary fairness 1265 (#1022, #1264) 1267 B.11. Since draft-ietf-httpbis-priority-00 1269 * Move text around (#1217, #1218) 1271 * Editorial change to the default urgency. The value is 3, which 1272 was always the intent of previous changes. 1274 B.12. Since draft-kazuho-httpbis-priority-04 1276 * Minimize semantics of Urgency levels (#1023, #1026) 1278 * Reduce guidance about how intermediary implements merging priority 1279 signals (#1026) 1281 * Remove mention of CDN-Loop (#1062) 1283 * Editorial changes 1285 * Make changes due to WG adoption 1287 * Removed outdated Consideration (#118) 1289 B.13. Since draft-kazuho-httpbis-priority-03 1291 * Changed numbering from [-1,6] to [0,7] (#78) 1293 * Replaced priority scheme negotiation with HTTP/2 priority 1294 deprecation (#100) 1296 * Shorten parameter names (#108) 1298 * Expand on considerations (#105, #107, #109, #110, #111, #113) 1300 B.14. Since draft-kazuho-httpbis-priority-02 1302 * Consolidation of the problem statement (#61, #73) 1304 * Define SETTINGS_PRIORITIES for negotiation (#58, #69) 1306 * Define PRIORITY_UPDATE frame for HTTP/2 and HTTP/3 (#51) 1308 * Explain fairness issue and mitigations (#56) 1310 B.15. Since draft-kazuho-httpbis-priority-01 1312 * Explain how reprioritization might be supported. 1314 B.16. Since draft-kazuho-httpbis-priority-00 1316 * Expand urgency levels from 3 to 8. 1318 Authors' Addresses 1320 Kazuho Oku 1321 Fastly 1323 Email: kazuhooku@gmail.com 1325 Lucas Pardue 1326 Cloudflare 1328 Email: lucaspardue.24.7@gmail.com