idnits 2.17.1 draft-ietf-dmm-requirements-17.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 == Line 955 has weird spacing: '...orkshop on Se...' == Line 978 has weird spacing: '...ference on Fu...' -- The document date (June 5, 2014) is 3169 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC6632' is defined on line 897, but no explicit reference was found in the text == Unused Reference: 'I-D.bhandari-dhc-class-based-prefix' is defined on line 910, but no explicit reference was found in the text ** Obsolete normative reference: RFC 3164 (Obsoleted by RFC 5424) -- Unexpected draft version: The latest known version of draft-wakikawa-netext-pmip-cp-up-separation is -00, but you're referring to -03. Summary: 1 error (**), 0 flaws (~~), 5 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group H. Chan (Ed.) 3 Internet-Draft Huawei Technologies 4 Intended status: Informational D. Liu 5 Expires: December 7, 2014 China Mobile 6 P. Seite 7 Orange 8 H. Yokota 9 KDDI Lab 10 J. Korhonen 11 Broadcom Communications 12 June 5, 2014 14 Requirements for Distributed Mobility Management 15 draft-ietf-dmm-requirements-17 17 Abstract 19 This document defines the requirements for Distributed Mobility 20 Management (DMM) at the network layer. The hierarchical structure in 21 traditional wireless networks has led primarily to centrally deployed 22 mobility anchors. As some wireless networks are evolving away from 23 the hierarchical structure, it can be useful to have a distributed 24 model for mobility management in which traffic does not need to 25 traverse centrally deployed mobility anchors far from the optimal 26 route. The motivation and the problems addressed by each requirement 27 are also described. 29 Requirements Language 31 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 32 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 33 document are to be interpreted as described in RFC 2119 [RFC2119]. 35 Status of this Memo 37 This Internet-Draft is submitted in full conformance with the 38 provisions of BCP 78 and BCP 79. 40 Internet-Drafts are working documents of the Internet Engineering 41 Task Force (IETF). Note that other groups may also distribute 42 working documents as Internet-Drafts. The list of current Internet- 43 Drafts is at http://datatracker.ietf.org/drafts/current/. 45 Internet-Drafts are draft documents valid for a maximum of six months 46 and may be updated, replaced, or obsoleted by other documents at any 47 time. It is inappropriate to use Internet-Drafts as reference 48 material or to cite them other than as "work in progress." 49 This Internet-Draft will expire on December 7, 2014. 51 Copyright Notice 53 Copyright (c) 2014 IETF Trust and the persons identified as the 54 document authors. All rights reserved. 56 This document is subject to BCP 78 and the IETF Trust's Legal 57 Provisions Relating to IETF Documents 58 (http://trustee.ietf.org/license-info) in effect on the date of 59 publication of this document. Please review these documents 60 carefully, as they describe your rights and restrictions with respect 61 to this document. Code Components extracted from this document must 62 include Simplified BSD License text as described in Section 4.e of 63 the Trust Legal Provisions and are provided without warranty as 64 described in the Simplified BSD License. 66 Table of Contents 68 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 69 2. Conventions used in this document . . . . . . . . . . . . . . 5 70 2.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 5 71 3. Centralized versus distributed mobility management . . . . . . 7 72 3.1. Centralized mobility management . . . . . . . . . . . . . 7 73 3.2. Distributed mobility management . . . . . . . . . . . . . 8 74 4. Problem Statement . . . . . . . . . . . . . . . . . . . . . . 9 75 5. Requirements . . . . . . . . . . . . . . . . . . . . . . . . . 11 76 6. Security Considerations . . . . . . . . . . . . . . . . . . . 17 77 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 17 78 8. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 17 79 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 20 80 9.1. Normative References . . . . . . . . . . . . . . . . . . . 20 81 9.2. Informative References . . . . . . . . . . . . . . . . . . 21 82 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 23 84 1. Introduction 86 In the past decade a fair number of network-layer mobility protocols 87 have been standardized [RFC6275] [RFC5944] [RFC5380] [RFC6301] 88 [RFC5213]. Although these protocols differ in terms of functions and 89 associated message formats, they all employ a mobility anchor to 90 allow a mobile node to remain reachable after it has moved to a 91 different network. The anchor point, among other tasks, ensures 92 connectivity by forwarding packets destined to, or sent from, the 93 mobile node. It is a centrally deployed mobility anchor in the sense 94 that the deployed architectures today have a small number of these 95 anchors and the traffic of millions of mobile nodes in an operator 96 network are typically managed by the same anchor. Such a mobility 97 anchor may still have to reside in the subscriber's provider network 98 even when the subscriber is roaming to a visited network, in order 99 that certain functions such as charging and billing can be performed 100 more readily by the provider's network. An example provider network 101 is a Third Generation Partnership Project (3GPP) network. 103 Distributed mobility management (DMM) is an alternative to the above 104 centralized deployment. The background behind the interests to study 105 DMM are primarily in the following. 107 (1) Mobile users are, more than ever, consuming Internet content 108 including that of local Content Delivery Networks (CDNs). Such 109 traffic imposes new requirements on mobile core networks for 110 data traffic delivery. To prevent exceeding the available core 111 network capacity, service providers need to implement new 112 strategies such as selective IPv4 traffic offload (e.g., 113 [RFC6909], 3GPP work items Local IP Access (LIPA) and Selected 114 IP Traffic Offload (SIPTO) [TS.23.401]) through alternative 115 access networks such as Wireless Local Area Network (WLAN) 116 [Paper-Mobile.Data.Offloading]. In addition, a gateway 117 selection mechanism takes the user proximity into account within 118 the Evolved Packet Core (EPC) [TS.29303]. Yet these mechanisms 119 were not pursued in the past owing to charging and billing 120 considerations which require solutions beyond the mobility 121 protocol. Consequently, assigning a gateway anchor node from a 122 visited network when roaming to the visited network has only 123 recently been done and is limited to voice services. 125 Both traffic offloading and CDN mechanisms could benefit from 126 the development of mobile architectures with fewer hierarchical 127 levels introduced into the data path by the mobility management 128 system. This trend of "flattening" the mobile networks works 129 best for direct communications among peers in the same 130 geographical area. Distributed mobility management in the 131 flattening mobile networks would anchor the traffic closer to 132 the point of attachment of the user. 134 (2) Today's mobile networks present service providers with new 135 challenges. Mobility patterns indicate that mobile nodes often 136 remain attached to the same point of attachment for considerable 137 periods of time [Paper-Locating.User]. Specific IP mobility 138 management support is not required for applications that launch 139 and complete their sessions while the mobile node is connected 140 to the same point of attachment. However, currently, IP 141 mobility support is designed for always-on operation, 142 maintaining all parameters of the context for each mobile 143 subscriber for as long as they are connected to the network. 144 This can result in a waste of resources and unnecessary costs 145 for the service provider. Infrequent node mobility coupled with 146 application intelligence suggest that mobility support could be 147 provided selectively such as in [I-D.bhandari-dhc-class-based- 148 prefix] and [I-D.korhonen-6man-prefix-properties], thus reducing 149 the amount of context maintained in the network. 151 DMM may distribute the mobility anchors in the data-plane in 152 flattening the mobility network such that the mobility anchors are 153 positioned closer to the user; ideally, mobility agents could be 154 collocated with the first-hop router. Facilitated by the 155 distribution of mobility anchors, it may be possible to selectively 156 use or not use mobility protocol support depending on whether such 157 support is needed or not. It can thus reduce the amount of state 158 information that must be maintained in various mobility agents of the 159 mobile network. It can then avoid the unnecessary establishment of 160 mechanisms to forward traffic from an old to a new mobility anchor. 162 This document compares distributed mobility management with 163 centralized mobility management in Section 3. The problems that can 164 be addressed with DMM are summarized in Section 4. The mandatory 165 requirements as well as the optional requirements for network-layer 166 distributed mobility management are given in Section 5. Finally, 167 security considerations are discussed in Section 6. 169 The problem statement and the use cases [I-D.yokota-dmm-scenario] can 170 be found in [Paper-Distributed.Mobility.Review]. 172 2. Conventions used in this document 174 2.1. Terminology 176 All the general mobility-related terms and their acronyms used in 177 this document are to be interpreted as defined in the Mobile IPv6 178 base specification [RFC6275], in the Proxy mobile IPv6 specification 180 [RFC5213], and in Mobility Related Terminology [RFC3753]. These 181 terms include the following: mobile node (MN), correspondent node 182 (CN), and home agent (HA) as per [RFC6275]; local mobility anchor 183 (LMA) and mobile access gateway (MAG) as per [RFC5213], and context 184 as per [RFC3753]. 186 In addition, this draft introduces the following terms. 188 Centrally deployed mobility anchors 190 refer to the mobility management deployments in which there are 191 very few mobility anchors and the traffic of millions of mobile 192 nodes in an operator network are managed by the same anchor. 194 Centralized mobility management 196 makes use of centrally deployed mobility anchors. 198 Distributed mobility management 200 is not centralized so that traffic does not need to traverse 201 centrally deployed mobility anchors far from the optimal route. 203 Hierarchical mobile network 205 has a hierarchy of network elements arranged into multiple 206 hierarchical levels which are introduced into the data path by the 207 mobility management system. 209 Flattening mobile network 211 refers to the hierarchical mobile network which is going through 212 the trend of reducing its number of hierarchical levels. 214 Flatter mobile network 216 has fewer hierarchical levels compared to a hierarchical mobile 217 network. 219 Mobility context 221 is the collection of information required to provide mobility 222 management support for a given mobile node. 224 3. Centralized versus distributed mobility management 226 Mobility management is needed because the IP address of a mobile node 227 may change as the node moves. Mobility management functions may be 228 implemented at different layers of the protocol stack. At the IP 229 (network) layer, mobility management can be client-based or network- 230 based. 232 An IP-layer mobility management protocol is typically based on the 233 principle of distinguishing between a session identifier and a 234 forwarding address and maintaining a mapping between the two. In 235 Mobile IP, the new IP address of the mobile node after the node has 236 moved is the forwarding address, whereas the original IP address 237 before the mobile node moves serves as the session identifier. The 238 location management (LM) information is kept by associating the 239 forwarding address with the session identifier. Packets addressed to 240 the session identifier will first route to the original network which 241 re-directs them using the forwarding address to deliver to the 242 session. Re-directing packets this way can result in long routes. 243 An existing optimization routes directly using the forwarding address 244 of the host, and such is a host-based solution. 246 The next two subsections explain centralized and distributed mobility 247 management functions in the network. 249 3.1. Centralized mobility management 251 In centralized mobility management, the location information in terms 252 of a mapping between the session identifier and the forwarding 253 address is kept at a single mobility anchor, and packets destined to 254 the session identifier are forwarded via this anchor. In other 255 words, such mobility management systems are centralized in both the 256 control plane and the data plane (mobile node IP traffic). 258 Many existing mobility management deployments make use of centralized 259 mobility anchoring in a hierarchical network architecture, as shown 260 in Figure 1. Examples are the home agent (HA) and local mobility 261 anchor (LMA) serving as the anchors for the mobile node (MN) and 262 Mobile Access Gateway (MAG) in Mobile IPv6 [RFC6275] and in Proxy 263 Mobile IPv6 [RFC5213] respectively. Cellular networks such as the 264 3GPP General Packet Radio System (GPRS) networks and 3GPP Evolved 265 Packet System (EPS) networks employ centralized mobility management 266 too. In the 3GPP GPRS network, the Gateway GPRS Support Node (GGSN), 267 Serving GPRS Support Node (SGSN) and Radio Network Controller (RNC) 268 constitute a hierarchy of anchors. In the 3GPP EPS network, the 269 Packet Data Network Gateway (P-GW) and Serving Gateway (S-GW) 270 constitute another hierarchy of anchors. 272 3GPP GPRS 3GPP EPS MIP/PMIP 273 +------+ +------+ +------+ 274 | GGSN | | P-GW | |HA/LMA| 275 +------+ +------+ +------+ 276 /\ /\ /\ 277 / \ / \ / \ 278 / \ / \ / \ 279 / \ / \ / \ 280 / \ / \ / \ 281 / \ / \ / \ 282 / \ / \ / \ 283 +------+ +------+ +------+ +------+ +------+ +------+ 284 | SGSN | | SGSN | | S-GW | | S-GW | |MN/MAG| |MN/MAG| 285 +------+ +------+ +------+ +------+ +------+ +------+ 286 /\ /\ 287 / \ / \ 288 / \ / \ 289 +---+ +---+ +---+ +---+ 290 |RNC| |RNC| |RNC| |RNC| 291 +---+ +---+ +---+ +---+ 293 Figure 1. Centralized mobility management. 295 3.2. Distributed mobility management 297 Mobility management functions may also be distributed in the data 298 plane to multiple networks as shown in Figure 2, so that a mobile 299 node in any of these networks may be served by a nearby function with 300 appropriate forwarding management (FM) capability. 302 +------+ +------+ +------+ +------+ 303 | FM | | FM | | FM | | FM | 304 +------+ +------+ +------+ +------+ 305 | 306 +----+ 307 | MN | 308 +----+ 310 Figure 2. Distributed mobility management. 312 DMM is distributed in the data plane, whereas the control plane may 313 either be centralized or distributed [I-D.yokota-dmm-scenario]. The 314 former case implicitly assumes separation of data and control planes 315 as described in [I-D.wakikawa-netext-pmip-cp-up-separation]. While 316 mobility management can be distributed, it is not necessary for other 317 functions such as subscription management, subscription database, and 318 network access authentication to be similarly distributed. 320 A distributed mobility management scheme for a flattening mobile 321 network consisting of access nodes is proposed in [Paper- 322 Distributed.Dynamic.Mobility]. Its benefits over centralized 323 mobility management have been shown through simulations [Paper- 324 Distributed.Centralized.Mobility]. Moreover, the (re)use and 325 extension of existing protocols in the design of both fully 326 distributed mobility management [Paper-Migrating.Home.Agents] [Paper- 327 Distributed.Mobility.SAE] and partially distributed mobility 328 management [Paper-Distributed.Mobility.PMIP] [Paper- 329 Distributed.Mobility.MIP] have been reported in the literature. 330 Therefore, before designing new mobility management protocols for a 331 future distributed architecture, it is recommended to first consider 332 whether existing mobility management protocols can be extended. 334 4. Problem Statement 336 The problems that can be addressed with DMM are summarized in the 337 following: 339 PS1: Non-optimal routes 341 Forwarding via a centralized anchor often results in non- 342 optimal routes, thereby increasing the end-to-end delay. The 343 problem is manifested, for example, when accessing a nearby 344 server or servers of a Content Delivery Network (CDN), or when 345 receiving locally available IP multicast or sending IP 346 multicast packets. (Existing route optimization is only a 347 host-based solution. On the other hand, localized routing with 348 PMIPv6 [RFC6705] addresses only a part of the problem where 349 both the MN and the correspondent node (CN) are attached to the 350 same MAG, and it is not applicable when the CN does not behave 351 like an MN.) 353 PS2: Divergence from other evolutionary trends in network 354 architectures such as distribution of content delivery. 356 Mobile networks have generally been evolving towards a flatter 357 and flatter network. Centralized mobility management, which is 358 non-optimal with a flatter network architecture, does not 359 support this evolution. 361 PS3: Lack of scalability of centralized tunnel management and 362 mobility context maintenance 364 Setting up tunnels through a central anchor and maintaining 365 mobility context for each MN usually requires more concentrated 366 resources in a centralized design, thus reducing scalability. 368 Distributing the tunnel maintenance function and the mobility 369 context maintenance function among different network entities 370 with proper signaling protocol design can avoid increasing the 371 concentrated resources with an increasing number of MNs. 373 PS4: Single point of failure and attack 375 Centralized anchoring designs may be more vulnerable to single 376 points of failures and attacks than a distributed system. The 377 impact of a successful attack on a system with centralized 378 mobility management can be far greater as well. 380 PS5: Unnecessary mobility support to clients that do not need it 382 IP mobility support is usually provided to all MNs. Yet it is 383 not always required, and not every parameter of mobility 384 context is always used. For example, some applications or 385 nodes do not need a stable IP address during a handover to 386 maintain session continuity. Sometimes, the entire application 387 session runs while the MN does not change the point of 388 attachment. Besides, some sessions, e.g., SIP-based sessions, 389 can handle mobility at the application layer and hence do not 390 need IP mobility support; it is then unnecessary to provide IP 391 mobility support for such sessions. 393 PS6: Mobility signaling overhead with peer-to-peer communication 395 Wasting resources when mobility signaling (e.g., maintenance of 396 the tunnel, keep alive signaling, etc.) is not turned off for 397 peer-to-peer communication. 399 PS7: Deployment with multiple mobility solutions 401 There are already many variants and extensions of MIP as well 402 mobility solutions at other layers. Deployment of new mobility 403 management solutions can be challenging, and debugging 404 difficult, when they co-exist with solutions already deployed 405 in the field. 407 PS8: Duplicate multicast traffic 409 IP multicast distribution over architectures using IP mobility 410 solutions (e.g., [RFC6224]) may lead to convergence of 411 duplicated multicast subscriptions towards the downstream 412 tunnel entity (e.g., MAG in PMIPv6). Concretely, when 413 multicast subscription for individual mobile nodes is coupled 414 with mobility tunnels (e.g., PMIPv6 tunnel), duplicate 415 multicast subscription(s) is prone to be received through 416 different upstream paths. This problem may also exist or be 417 more severe in a distributed mobility environment. 419 5. Requirements 421 After comparing distributed mobility management against centralized 422 deployment in Section 3 and describing the problems in Section 4, 423 this section identifies the following requirements: 425 REQ1: Distributed mobility management 427 IP mobility, network access and forwarding solutions provided 428 by DMM MUST enable traffic to avoid traversing single mobility 429 anchor far from the optimal route. 431 This requirement on distribution is in the data plane only. 432 It does not impose constraints on whether the control plane 433 should be distributed or centralized. However, if the control 434 plane is centralized while the data plane is distributed, it 435 is implicit that the control plane and data plane need to 436 separate (Section 3.2). 438 Motivation: This requirement is motivated by current trends in 439 network evolution: (a) it is cost- and resource-effective to 440 cache contents, and the caching (e.g., CDN) servers are 441 distributed so that each user in any location can be close to 442 one of the servers; (b) the significantly larger number of 443 mobile nodes and flows call for improved scalability; (c) 444 single points of failure are avoided in a distributed system; 445 (d) threats against centrally deployed anchors, e.g., home 446 agent and local mobility anchor, are mitigated in a 447 distributed system. 449 This requirement addresses the problems PS1, PS2, PS3, and PS4 450 described in Section 4. 452 REQ2: Bypassable network-layer mobility support for each application 453 session 455 DMM solutions MUST enable network-layer mobility but it MUST 456 be possible for any individual active application session 457 (flow) to not use it. Mobility support is needed, for 458 example, when a mobile host moves and an application cannot 459 cope with a change in the IP address. Mobility support is 460 also needed when a mobile router changes its IP address as it 461 moves together with a host and, in the presence of ingress 462 filtering, an application in the host is interrupted. However 463 mobility support at the network-layer is not always needed; a 464 mobile node can often be stationary, and mobility support can 465 also be provided at other layers. It is then not always 466 necessary to maintain a stable IP address or prefix for an 467 active application session. 469 Different active sessions can also differ in whether network- 470 layer mobility support is needed. IP mobility, network access 471 and forwarding solutions provided by DMM MUST then enable the 472 possibility of independent handling for each application 473 session of a user or mobile device. 475 The handling of mobility management to the granularity of an 476 individual session of a user/device SHOULD need proper session 477 identification in addition to user/device identification. 479 Motivation: The motivation of this requirement is to enable 480 more efficient forwarding and more efficient use of network 481 resources by selecting an IP address or prefix according to 482 whether mobility support is needed and by not maintaining 483 context at the mobility anchor when there is no such need. 485 This requirement addresses the problems PS5 and PS6 described 486 in Section 4. 488 REQ3: IPv6 deployment 490 DMM solutions SHOULD target IPv6 as the primary deployment 491 environment and SHOULD NOT be tailored specifically to support 492 IPv4, in particular in situations where private IPv4 addresses 493 and/or NATs are used. 495 Motivation: This requirement conforms to the general 496 orientation of IETF work. DMM deployment is foreseen in mid- 497 to long-term horizon, when IPv6 is expected to be far more 498 common than today. 500 This requirement avoids the unnecessarily complexity in 501 solving the problems in Section 4 for IPv4, which will not be 502 able to use some of the IPv6-specific features. 504 REQ4: Existing mobility protocols 506 A DMM solution MUST first consider reusing and extending IETF- 507 standardized protocols before specifying new protocols. 509 Motivation: Reuse of existing IETF work is more efficient and 510 less error-prone. 512 This requirement attempts to avoid the need of new protocols 513 development and therefore their potential problems of being 514 time-consuming and error-prone. 516 REQ5: Coexistence with deployed networks/hosts and operability 517 across different networks 519 A DMM solution may require loose, tight or no integration into 520 existing mobility protocols and host IP stack. Regardless of 521 the integration level, DMM implementations MUST be able to 522 coexist with existing network deployments, end hosts and 523 routers that may or may not implement existing mobility 524 protocols. Furthermore, a DMM solution SHOULD work across 525 different networks, possibly operated as separate 526 administrative domains, when the needed mobility management 527 signaling, forwarding, and network access are allowed by the 528 trust relationship between them. 530 Motivation: (a) to preserve backwards compatibility so that 531 existing networks and hosts are not affected and continue to 532 function as usual, and (b) enable inter-domain operation if 533 desired. 535 This requirement addresses the problem PS7 described in 536 Section 4. 538 REQ6: Operation and Management considerations. 540 A DMM solution needs to consider configuring a device, 541 monitoring the current operational state of a device, 542 responding to events that impact the device, possibly by 543 modifying the configuration and storing the data in a format 544 that can be analyzed later. Different management protocols 545 are available. For example: 547 (a) SNMP [RFC1157] with definition of standardized management 548 information base MIB objects for DMM, that allows 549 monitoring traffic steering in a consistent manner across 550 different devices, 552 (b) NETCONF [RFC6241] with definition of standardized YANG 553 [RFC6020] modules for DMM to achieve a standardized 554 configuration, 556 (c) syslog [RFC3164] which is a one-way protocol allowing a 557 device to report significant events to a log analyzer in 558 a network management system. 560 (d) IP Flow Information Export (IPFIX) Protocol, which serves 561 as a means for transmitting traffic flow information over 562 the network [RFC7011], with a formal description of IPFIX 563 Information Elements [RFC7012]. 565 It is not the goal of the requirements document to impose 566 which management protocol(s) should be used. An inventory of 567 the management protocols and data models is covered in RFC 568 6632. 570 The following lists the operation and management 571 considerations required for a DMM solution; the list may not 572 be exhaustive and may be expanded according to the needs of 573 the solutions: 575 A DMM solution MUST describe in what environment and how it 576 can be scalably deployed and managed. 578 A DMM solution MUST support mechanisms to test if the DMM 579 solution is working properly. For example, when a DMM 580 solution employs traffic indirection to support a mobility 581 session, implementations MUST support mechanisms to test that 582 the appropriate traffic indirection operations are in place, 583 including the setup of traffic indirection and the subsequent 584 teardown of the indirection to release the associated network 585 resources when the mobility session has closed. 587 A DMM solution SHOULD expose the operational state of DMM to 588 the administrators of the DMM entities. For example, when a 589 DMM solution employs separation between session identifier and 590 forwarding address, it should expose the association between 591 them. 593 When flow mobility is supported by a DMM solution, the 594 solution SHOULD support means to correlate the flow routing 595 policies and the observed forwarding actions. 597 A DMM solution SHOULD support mechanisms to check the liveness 598 of forwarding path. If the DMM solution sends periodic update 599 refresh messages to configure the forwarding path, the refresh 600 period SHOULD be configurable and a reasonable default 601 configuration value proposed. Information collected can be 602 logged or made available with protocols such as SNMP 603 [RFC1157], NETCONF [RFC6241], IPFIX [RFC7011], or syslog 604 [RFC3164]. 606 A DMM solution MUST provide fault management and monitoring 607 mechanisms to manage situations where update of the mobility 608 session or the data path fails. The system must also be able 609 to handle situations where a mobility anchor with ongoing 610 mobility sessions fails. 612 A DMM solution SHOULD be able to monitor usage of DMM 613 protocol. When a DMM solution uses an existing protocol, the 614 techniques already defined for that protocol SHOULD be used to 615 monitor the DMM operation. When these techniques are 616 inadequate, new techniques MUST be developed. 618 In particular, the DMM solution SHOULD 620 (a) be able to monitor the number of mobility sessions per 621 user as well as their average duration. 623 (b) provide indication on DMM performance such as 625 1 the handover delay which includes the time necessary 626 to re-establish the forwarding path when the point of 627 attachment changes, 629 2 the protocol reactivity which is the time between 630 handover events such as the attachment to a new access 631 point and the completion of the mobility session 632 update. 634 (c) provide means to measure the signaling cost of the DMM 635 protocol. 637 (d) if tunneling is used for traffic redirection, monitor 639 1 the number of tunnels, 641 2 their transmission and reception information, 643 3 the used encapsulation method and overhead 645 4 the security used at a node level. 647 DMM solutions SHOULD support standardized configuration with 648 NETCONF [RFC6241], using YANG [RFC6020] modules, which SHOULD 649 be created for DMM when needed for such configuration. 650 However, if a DMM solution creates extensions to MIPv6 or 651 PMIPv6, the allowed addition of the definition of management 652 information base (MIB) objects to MIPv6 MIB [RFC4295] or 653 PMIPv6 MIB [RFC6475] needed for the control and monitoring of 654 the protocol extensions SHOULD be limited to read-only 655 objects. 657 Motivation: A DMM solution that is designed from the beginning 658 for operability and manageability can avoid difficulty or 659 incompatibility to implement efficient operations and 660 management solutions. 662 These requirements avoid DMM designs that make operations and 663 management difficult or costly. 665 REQ7: Security considerations 667 A DMM solution MUST support any security protocols and 668 mechanisms needed to secure the network and to make continuous 669 security improvements. In addition, with security taken into 670 consideration early in the design, a DMM solution MUST NOT 671 introduce new security risks, or amplify existing security 672 risks, that cannot be mitigated by existing security protocols 673 and mechanisms. 675 Motivation: Various attacks such as impersonation, denial of 676 service, man-in-the-middle attacks, and so on, may be launched 677 in a DMM deployment. For instance, an illegitimate node may 678 attempt to access a network providing DMM. Another example is 679 that a malicious node can forge a number of signaling messages 680 thus redirecting traffic from its legitimate path. 681 Consequently, the specific node or nodes to which the traffic 682 is redirected may be under a denial of service attack, whereas 683 other nodes do not receive their traffic. Accordingly, 684 security mechanisms/protocols providing access control, 685 integrity, authentication, authorization, confidentiality, 686 etc. should be used to protect the DMM entities as they are 687 already used to protect against existing networks and existing 688 mobility protocols defined in IETF. Yet if a candidate DMM 689 solution is such that even the proper use of these existing 690 security mechanisms/protocols are unable to provide sufficient 691 security protection, that candidate DMM solution is causing 692 uncontrollable security problems. 694 This requirement prevents a DMM solution from introducing 695 uncontrollable problems of potentially insecure mobility 696 management protocols which make deployment infeasible because 697 platforms conforming to the protocols are at risk for data 698 loss and numerous other dangers, including financial harm to 699 the users. 701 REQ8: Multicast considerations 703 DMM SHOULD enable multicast solutions to be developed to avoid 704 network inefficiency in multicast traffic delivery. 706 Motivation: Existing multicast deployment have been introduced 707 after completing the design of the reference mobility 708 protocol, often leading to network inefficiency and non- 709 optimal forwarding for the multicast traffic. Instead DMM 710 should consider multicast early so that the multicast 711 solutions can better consider efficiency nature in the 712 multicast traffic delivery (such as duplicate multicast 713 subscriptions towards the downstream tunnel entities). The 714 multicast solutions should then avoid restricting the 715 management of all IP multicast traffic to a single host 716 through a dedicated (tunnel) interface on multicast-capable 717 access routers. 719 This requirement addresses the problems PS1 and PS8 described 720 in Section 4. 722 6. Security Considerations 724 Please refer to the discussion under Security requirement in Section 725 5. 727 7. IANA Considerations 729 None 731 8. Contributors 733 This requirements document is a joint effort among numerous 734 participants working in a team. Valuable comments and suggestions in 735 various reviews from the following area directors and IESG members 736 have also contributed to much improvements: Russ Housley, Catherine 737 Meadows, Adrian Farrel, Barry Leiba, Alissa Cooper, Ted Lemon, Brian 738 Haberman, Stephen Farrell, Joel Jaeggli, Alia Atlas, and Benoit 739 Claise. In addition to the authors, each of the following has made 740 very significant and important contributions to the working group 741 draft in this work: 743 Charles E. Perkins 744 Huawei Technologies 745 Email: charliep@computer.org 746 Melia Telemaco 747 Alcatel-Lucent Bell Labs 748 Email: telemaco.melia@googlemail.com 750 Elena Demaria 751 Telecom Italia 752 via G. Reiss Romoli, 274, TORINO, 10148, Italy 753 Email: elena.demaria@telecomitalia.it 755 Jong-Hyouk Lee 756 Sangmyung University, Korea 757 Email: jonghyouk@smu.ac.kr 759 Kostas Pentikousis 760 EICT GmbH 761 Email: k.pentikousis@eict.de 763 Tricci So 764 ZTE 765 Email: tso@zteusa.com 767 Carlos J. Bernardos 768 Universidad Carlos III de Madrid 769 Av. Universidad, 30, Leganes, Madrid 28911, Spain 770 Email: cjbc@it.uc3m.es 772 Peter McCann 773 Huawei Technologies 774 Email: Peter.McCann@huawei.com 776 Seok Joo Koh 777 Kyungpook National University, Korea 778 Email: sjkoh@knu.ac.kr 780 Wen Luo 781 ZTE 782 No.68, Zijinhua RD,Yuhuatai District, Nanjing, Jiangsu 210012, China 783 Email: luo.wen@zte.com.cn 785 Sri Gundavelli 786 Cisco 787 sgundave@cisco.com 789 Hui Deng 790 China Mobile 791 Email: denghui@chinamobile.com 793 Marco Liebsch 794 NEC Laboratories Europe 795 Email: liebsch@neclab.eu 797 Carl Williams 798 MCSR Labs 799 Email: carlw@mcsr-labs.org 801 Seil Jeon 802 Instituto de Telecomunicacoes, Aveiro 803 Email: seiljeon@av.it.pt 805 Sergio Figueiredo 806 Universidade de Aveiro 807 Email: sfigueiredo@av.it.pt 809 Stig Venaas 810 Email: stig@venaas.com 812 Luis Miguel Contreras Murillo 813 Telefonica I+D 814 Email: lmcm@tid.es 816 Juan Carlos Zuniga 817 InterDigital 818 Email: JuanCarlos.Zuniga@InterDigital.com 820 Alexandru Petrescu 821 Email: alexandru.petrescu@gmail.com 823 Georgios Karagiannis 824 University of Twente 825 Email: g.karagiannis@utwente.nl 827 Julien Laganier 828 Juniper 829 Email: julien.ietf@gmail.com 831 Wassim Michel Haddad 832 Ericsson 833 Email: Wassim.Haddad@ericsson.com 835 Dirk von Hugo 836 Deutsche Telekom Laboratories 837 Email: Dirk.von-Hugo@telekom.de 839 Ahmad Muhanna 840 Award Solutions 841 Email: asmuhanna@yahoo.com 842 Byoung-Jo Kim 843 ATT Labs 844 Email: macsbug@research.att.com 846 Hassan Ali-Ahmad 847 Orange 848 Email: hassan.aliahmad@orange.com 850 Alper Yegin 851 Samsung 852 Email: alper.yegin@partner.samsung.com 854 David Harrington 855 Effective Software 856 Email: ietfdbh@comcast.net 858 9. References 860 9.1. Normative References 862 [RFC1157] Case, J., Fedor, M., Schoffstall, M., and J. Davin, 863 "Simple Network Management Protocol (SNMP)", STD 15, 864 RFC 1157, May 1990. 866 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 867 Requirement Levels", BCP 14, RFC 2119, March 1997. 869 [RFC3164] Lonvick, C., "The BSD Syslog Protocol", RFC 3164, 870 August 2001. 872 [RFC3753] Manner, J. and M. Kojo, "Mobility Related Terminology", 873 RFC 3753, June 2004. 875 [RFC4295] Keeni, G., Koide, K., Nagami, K., and S. Gundavelli, 876 "Mobile IPv6 Management Information Base", RFC 4295, 877 April 2006. 879 [RFC5213] Gundavelli, S., Leung, K., Devarapalli, V., Chowdhury, K., 880 and B. Patil, "Proxy Mobile IPv6", RFC 5213, August 2008. 882 [RFC6020] Bjorklund, M., "YANG - A Data Modeling Language for the 883 Network Configuration Protocol (NETCONF)", RFC 6020, 884 October 2010. 886 [RFC6241] Enns, R., Bjorklund, M., Schoenwaelder, J., and A. 887 Bierman, "Network Configuration Protocol (NETCONF)", 888 RFC 6241, June 2011. 890 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 891 in IPv6", RFC 6275, July 2011. 893 [RFC6475] Keeni, G., Koide, K., Gundavelli, S., and R. Wakikawa, 894 "Proxy Mobile IPv6 Management Information Base", RFC 6475, 895 May 2012. 897 [RFC6632] Ersue, M. and B. Claise, "An Overview of the IETF Network 898 Management Standards", RFC 6632, June 2012. 900 [RFC7011] Claise, B., Trammell, B., and P. Aitken, "Specification of 901 the IP Flow Information Export (IPFIX) Protocol for the 902 Exchange of Flow Information", STD 77, RFC 7011, 903 September 2013. 905 [RFC7012] Claise, B. and B. Trammell, "Information Model for IP Flow 906 Information Export (IPFIX)", RFC 7012, September 2013. 908 9.2. Informative References 910 [I-D.bhandari-dhc-class-based-prefix] 911 Bhandari, S., Halwasia, G., Gundavelli, S., Deng, H., 912 Thiebaut, L., Korhonen, J., and I. Farrer, "DHCPv6 class 913 based prefix", draft-bhandari-dhc-class-based-prefix-05 914 (work in progress), July 2013. 916 [I-D.korhonen-6man-prefix-properties] 917 Korhonen, J., Patil, B., Gundavelli, S., Seite, P., and D. 918 Liu, "IPv6 Prefix Properties", 919 draft-korhonen-6man-prefix-properties-02 (work in 920 progress), July 2013. 922 [I-D.wakikawa-netext-pmip-cp-up-separation] 923 Wakikawa, R., Pazhyannur, R., Gundavelli, S., and C. 924 Perkins, "Separation of Control and User Plane for Proxy 925 Mobile IPv6", 926 draft-wakikawa-netext-pmip-cp-up-separation-03 (work in 927 progress), April 2014. 929 [I-D.yokota-dmm-scenario] 930 Yokota, H., Seite, P., Demaria, E., and Z. Cao, "Use case 931 scenarios for Distributed Mobility Management", 932 draft-yokota-dmm-scenario-00 (work in progress), 933 October 2010. 935 [Paper-Distributed.Centralized.Mobility] 936 Bertin, P., Bonjour, S., and J-M. Bonnin, "A Distributed 937 or Centralized Mobility", Proceedings of Global 938 Communications Conference (GlobeCom), December 2009. 940 [Paper-Distributed.Dynamic.Mobility] 941 Bertin, P., Bonjour, S., and J-M. Bonnin, "A Distributed 942 Dynamic Mobility Management Scheme Designed for Flat IP 943 Architectures", Proceedings of 3rd International 944 Conference on New Technologies, Mobility and Security 945 (NTMS), 2008. 947 [Paper-Distributed.Mobility.MIP] 948 Chan, H., "Distributed Mobility Management with Mobile 949 IP", Proceedings of IEEE International Communication 950 Conference (ICC) Workshop on Telecommunications: from 951 Research to Standards, June 2012. 953 [Paper-Distributed.Mobility.PMIP] 954 Chan, H., "Proxy Mobile IP with Distributed Mobility 955 Anchors", Proceedings of GlobeCom Workshop on Seamless 956 Wireless Mobility, December 2010. 958 [Paper-Distributed.Mobility.Review] 959 Chan, H., Yokota, H., Xie, J., Seite, P., and D. Liu, 960 "Distributed and Dynamic Mobility Management in Mobile 961 Internet: Current Approaches and Issues", Journal of 962 Communications, vol. 6, no. 1, pp. 4-15, February 2011. 964 [Paper-Distributed.Mobility.SAE] 965 Fisher, M., Anderson, F., Kopsel, A., Schafer, G., and M. 966 Schlager, "A Distributed IP Mobility Approach for 3G SAE", 967 Proceedings of the 19th International Symposium on 968 Personal, Indoor and Mobile Radio Communications (PIMRC), 969 2008. 971 [Paper-Locating.User] 972 Kirby, G., "Locating the User", Communication 973 International, 1995. 975 [Paper-Migrating.Home.Agents] 976 Wakikawa, R., Valadon, G., and J. Murai, "Migrating Home 977 Agents Towards Internet-scale Mobility Deployments", 978 Proceedings of the ACM 2nd CoNEXT Conference on Future 979 Networking Technologies, December 2006. 981 [Paper-Mobile.Data.Offloading] 982 Lee, K., Lee, J., Yi, Y., Rhee, I., and S. Chong, "Mobile 983 Data Offloading: How Much Can WiFi Deliver?", SIGCOMM 984 2010, 2010. 986 [RFC5380] Soliman, H., Castelluccia, C., ElMalki, K., and L. 987 Bellier, "Hierarchical Mobile IPv6 (HMIPv6) Mobility 988 Management", RFC 5380, October 2008. 990 [RFC5944] Perkins, C., "IP Mobility Support for IPv4, Revised", 991 RFC 5944, November 2010. 993 [RFC6224] Schmidt, T., Waehlisch, M., and S. Krishnan, "Base 994 Deployment for Multicast Listener Support in Proxy Mobile 995 IPv6 (PMIPv6) Domains", RFC 6224, April 2011. 997 [RFC6301] Zhu, Z., Wakikawa, R., and L. Zhang, "A Survey of Mobility 998 Support in the Internet", RFC 6301, July 2011. 1000 [RFC6705] Krishnan, S., Koodli, R., Loureiro, P., Wu, Q., and A. 1001 Dutta, "Localized Routing for Proxy Mobile IPv6", 1002 RFC 6705, September 2012. 1004 [RFC6909] Gundavelli, S., Zhou, X., Korhonen, J., Feige, G., and R. 1005 Koodli, "IPv4 Traffic Offload Selector Option for Proxy 1006 Mobile IPv6", RFC 6909, April 2013. 1008 [TS.23.401] 1009 3GPP, "General Packet Radio Service (GPRS) enhancements 1010 for Evolved Universal Terrestrial Radio Access Network 1011 (E-UTRAN) access", 3GPP TR 23.401 10.10.0, March 2013. 1013 [TS.29303] 1014 3GPP, "Domain Name System Procedures; Stage 3", 3GPP 1015 TR 23.303 11.2.0, September 2012. 1017 Authors' Addresses 1019 H Anthony Chan (editor) 1020 Huawei Technologies 1021 5340 Legacy Dr. Building 3, Plano, TX 75024, USA 1022 Email: h.a.chan@ieee.org 1024 Dapeng Liu 1025 China Mobile 1026 Unit2, 28 Xuanwumenxi Ave, Xuanwu District, Beijing 100053, China 1027 Email: liudapeng@chinamobile.com 1028 Pierrick Seite 1029 Orange 1030 4, rue du Clos Courtel, BP 91226, Cesson-Sevigne 35512, France 1031 Email: pierrick.seite@orange.com 1033 Hidetoshi Yokota 1034 KDDI Lab 1035 2-1-15 Ohara, Fujimino, Saitama, 356-8502 Japan 1036 Email: yokota@kddilabs.jp 1038 Jouni Korhonen 1039 Broadcom Communications 1040 Porkkalankatu 24, FIN-00180 Helsinki, Finland 1041 Email: jouni.nospam@gmail.com