idnits 2.17.1 draft-ietf-rtgwg-bgp-routing-large-dc-00.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 (August 14, 2014) is 3543 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-15) exists of draft-ietf-idr-add-paths-09 == Outdated reference: A later version (-07) exists of draft-ietf-idr-link-bandwidth-06 Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Routing Area Working Group P. Lapukhov 3 Internet-Draft Facebook 4 Intended status: Informational A. Premji 5 Expires: February 15, 2015 Arista Networks 6 J. Mitchell, Ed. 7 Microsoft Corporation 8 August 14, 2014 10 Use of BGP for routing in large-scale data centers 11 draft-ietf-rtgwg-bgp-routing-large-dc-00 13 Abstract 15 Some network operators build and operate data centers that support 16 over one hundred thousand servers. In this document, such data 17 centers are referred to as "large-scale" to differentiate them from 18 smaller infrastructures. Environments of this scale have a unique 19 set of network requirements with an emphasis on operational 20 simplicity and network stability. This document summarizes 21 operational experience in designing and operating large-scale data 22 centers using BGP as the only routing protocol. The intent is to 23 report on a proven and stable routing design that could be leveraged 24 by others in the industry. 26 Status of This Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at http://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on February 15, 2015. 43 Copyright Notice 45 Copyright (c) 2014 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (http://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 61 2. Network Design Requirements . . . . . . . . . . . . . . . . . 4 62 2.1. Bandwidth and Traffic Patterns . . . . . . . . . . . . . 4 63 2.2. CAPEX Minimization . . . . . . . . . . . . . . . . . . . 4 64 2.3. OPEX Minimization . . . . . . . . . . . . . . . . . . . . 5 65 2.4. Traffic Engineering . . . . . . . . . . . . . . . . . . . 5 66 2.5. Summarized Requirements . . . . . . . . . . . . . . . . . 5 67 3. Data Center Topologies Overview . . . . . . . . . . . . . . . 6 68 3.1. Traditional DC Topology . . . . . . . . . . . . . . . . . 6 69 3.2. Clos Network topology . . . . . . . . . . . . . . . . . . 7 70 3.2.1. Overview . . . . . . . . . . . . . . . . . . . . . . 7 71 3.2.2. Clos Topology Properties . . . . . . . . . . . . . . 8 72 3.2.3. Scaling the Clos topology . . . . . . . . . . . . . . 9 73 3.2.4. Managing the Size of Clos Topology Tiers . . . . . . 10 74 4. Data Center Routing Overview . . . . . . . . . . . . . . . . 10 75 4.1. Layer 2 Only Designs . . . . . . . . . . . . . . . . . . 11 76 4.2. Hybrid L2/L3 Designs . . . . . . . . . . . . . . . . . . 11 77 4.3. Layer 3 Only Designs . . . . . . . . . . . . . . . . . . 12 78 5. Routing Protocol Selection and Design . . . . . . . . . . . . 12 79 5.1. Choosing EBGP as the Routing Protocol . . . . . . . . . . 13 80 5.2. EBGP Configuration for Clos topology . . . . . . . . . . 14 81 5.2.1. Example ASN Scheme . . . . . . . . . . . . . . . . . 14 82 5.2.2. Private Use BGP ASNs . . . . . . . . . . . . . . . . 15 83 5.2.3. Prefix Advertisement . . . . . . . . . . . . . . . . 16 84 5.2.4. External Connectivity . . . . . . . . . . . . . . . . 17 85 5.2.5. Route Summarization at the Edge . . . . . . . . . . . 18 86 6. ECMP Considerations . . . . . . . . . . . . . . . . . . . . . 19 87 6.1. Basic ECMP . . . . . . . . . . . . . . . . . . . . . . . 19 88 6.2. BGP ECMP over Multiple ASNs . . . . . . . . . . . . . . . 20 89 6.3. Weighted ECMP . . . . . . . . . . . . . . . . . . . . . . 20 90 6.4. Consistent Hashing . . . . . . . . . . . . . . . . . . . 21 91 7. Routing Convergence Properties . . . . . . . . . . . . . . . 21 92 7.1. Fault Detection Timing . . . . . . . . . . . . . . . . . 21 93 7.2. Event Propagation Timing . . . . . . . . . . . . . . . . 22 94 7.3. Impact of Clos Topology Fan-outs . . . . . . . . . . . . 22 95 7.4. Failure Impact Scope . . . . . . . . . . . . . . . . . . 23 96 7.5. Routing Micro-Loops . . . . . . . . . . . . . . . . . . . 24 97 8. Additional Options for Design . . . . . . . . . . . . . . . . 25 98 8.1. Third-party Route Injection . . . . . . . . . . . . . . . 25 99 8.2. Route Summarization within Clos Topology . . . . . . . . 25 100 8.2.1. Collapsing Tier-1 Devices Layer . . . . . . . . . . . 25 101 8.2.2. Simple Virtual Aggregation . . . . . . . . . . . . . 27 102 8.3. ICMP Unreachable Message Masquerading . . . . . . . . . . 27 103 9. Security Considerations . . . . . . . . . . . . . . . . . . . 28 104 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 28 105 11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 28 106 12. References . . . . . . . . . . . . . . . . . . . . . . . . . 28 107 12.1. Normative References . . . . . . . . . . . . . . . . . . 28 108 12.2. Informative References . . . . . . . . . . . . . . . . . 29 109 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 31 111 1. Introduction 113 This document describes a practical routing design that can be used 114 in a large-scale data center ("DC") design. Such data centers, also 115 known as hyper-scale or warehouse-scale data-centers, have a unique 116 attribute of supporting over a hundred thousand servers. In order to 117 accommodate networks of this scale, operators are revisiting 118 networking designs and platforms to address this need. 120 The design presented in this document is based on operational 121 experience with data centers built to support large scale distributed 122 software infrastructure, such as a Web search engine. The primary 123 requirements in such an environment are operational simplicity and 124 network stability so that a small group of people can effectively 125 support a significantly sized network. 127 After experimentation and extensive testing, Microsoft chose to use 128 an end-to-end routed network infrastructure with External BGP (EBGP) 129 [RFC4271] as the only routing protocol for some of its DC 130 deployments. This is in contrast with more traditional DC designs, 131 which may use simple tree topologies and rely on extending Layer 2 132 domains across multiple network devices. This document elaborates on 133 the requirements that led to this design choice and presents details 134 of the EBGP routing design as well as explores ideas for further 135 enhancements. 137 This document first presents an overview of network design 138 requirements and considerations for large-scale data centers. Then 139 traditional hierarchical data center network topologies are 140 contrasted with Clos networks that are horizontally scaled out. This 141 is followed by arguments for selecting EBGP with a Clos topology as 142 the most appropriate routing protocol to meet the requirements and 143 the proposed design is described in detail. Finally, the document 144 reviews some additional considerations and design options. 146 2. Network Design Requirements 148 This section describes and summarizes network design requirements for 149 large-scale data centers. 151 2.1. Bandwidth and Traffic Patterns 153 The primary requirement when building an interconnection network for 154 large number of servers is to accommodate application bandwidth and 155 latency requirements. Until recently it was quite common to see the 156 majority of traffic entering and leaving the data center, commonly 157 referred to as "north-south" traffic. As a result, traditional 158 "tree" topologies were sufficient to accommodate such flows, even 159 with high oversubscription ratios between the layers of the network. 160 If more bandwidth was required, it was added by "scaling up" the 161 network elements, e.g. by upgrading the device's line-cards or 162 fabrics or replacing the device with one with higher port density. 164 Today many large-scale data centers host applications generating 165 significant amounts of server-to-server traffic, which does not 166 egress the DC, commonly referred to as "east-west" traffic. Examples 167 of such applications could be compute clusters such as Hadoop, 168 massive data replication between clusters needed by certain 169 applications, or virtual machine migrations. Scaling traditional 170 tree topologies to match these bandwidth demands becomes either too 171 expensive or impossible due to physical limitations, e.g. port 172 density in a switch. 174 2.2. CAPEX Minimization 176 The cost of the network infrastructure alone (CAPEX) constitutes 177 about 10-15% of total data center expenditure (see [GREENBERG2009]). 178 However, the absolute cost is significant, and hence there is a need 179 to constantly drive down the cost of individual network elements. 180 This can be accomplished in two ways: 182 o Unifying all network elements, preferably using the same hardware 183 type or even the same device. This allows for volume pricing on 184 bulk purchases. 186 o Driving costs down using competitive pressures, by introducing 187 multiple network equipment vendors. 189 In order to allow for good vendor diversity it is important to 190 minimize the software feature requirements for the network elements. 192 This strategy provides maximum flexibility of vendor equipment 193 choices while enforcing interoperability using open standards. 195 2.3. OPEX Minimization 197 Operating large-scale infrastructure could be expensive, provided 198 that larger amount of elements will statistically fail more often. 199 Having a simpler design and operating using a limited software 200 feature-set minimizes software issue related failures. 202 An important aspect of OPEX minimization is reducing size of failure 203 domains in the network. Ethernet networks are known to be 204 susceptible to broadcast or unicast traffic storms that have dramatic 205 impact on network performance and availability. The use of a fully 206 routed design significantly reduces the size of the data-plane 207 failure domains - i.e. limits them to the lowest level in the network 208 hierarchy. However, such designs introduce the problem of 209 distributed control-plane failures. This observation calls for 210 simpler control-plane protocols that are expected to have less 211 chances of network meltdown. Minimizing software feature 212 requirements as described in the CAPEX section above also reduces 213 testing and training requirements. 215 2.4. Traffic Engineering 217 In any data center, application load-balancing is a critical function 218 performed by network devices. Traditionally, load-balancers are 219 deployed as dedicated devices in the traffic forwarding path. The 220 problem arises in scaling load-balancers under growing traffic 221 demand. A preferable solution would be able to scale load-balancing 222 layer horizontally, by adding more of the uniform nodes and 223 distributing incoming traffic across these nodes. In situation like 224 this, an ideal choice would be to use network infrastructure itself 225 to distribute traffic across a group of load-balancers. The 226 combination of Anycast prefix advertisement [RFC4786] and Equal Cost 227 Multipath (ECMP) functionality can be used to accomplish this goal. 228 To allow for more granular load-distribution, it is beneficial for 229 the network to support the ability to perform controlled per-hop 230 traffic engineering. For example, it is beneficial to directly 231 control the ECMP next-hop set for Anycast prefixes at every level of 232 network hierarchy. 234 2.5. Summarized Requirements 236 This section summarizes the list of requirements outlined in the 237 previous sections: 239 o REQ1: Select a topology that can be scaled "horizontally" by 240 adding more links and network devices of the same type without 241 requiring upgrades to the network elements themselves. 243 o REQ2: Define a narrow set of software features/protocols supported 244 by a multitude of networking equipment vendors. 246 o REQ3: Choose a routing protocol that has a simple implementation 247 in terms of programming code complexity and ease of operational 248 support. 250 o REQ4: Minimize the failure domain of equipment or protocol issues 251 as much as possible. 253 o REQ5: Allow for traffic engineering, preferably via explicit 254 control of the routing prefix next-hop using built-in protocol 255 mechanics. 257 3. Data Center Topologies Overview 259 This section provides an overview of two general types of data center 260 designs - hierarchical (also known as tree based) and Clos based 261 network designs. 263 3.1. Traditional DC Topology 265 In the networking industry, a common design choice for data centers 266 typically look like a (upside-down) tree with redundant uplinks and 267 three layers of hierarchy namely core, aggregation/distribution and 268 access layers (see Figure 1). To accommodate bandwidth demands, each 269 higher layer, from server towards DC egress or WAN, has higher port 270 density and bandwidth capacity where the core functions as the 271 "trunk" of the tree based design. To keep terminology uniform and 272 for comparison with other designs, in this document these layers will 273 be referred to as Tier-1, Tier-2 and Tier-3 "tiers" instead of Core, 274 Aggregation or Access layers. 276 +------+ +------+ 277 | | | | 278 | |--| | Tier-1 279 | | | | 280 +------+ +------+ 281 | | | | 282 +---------+ | | +----------+ 283 | +-------+--+------+--+-------+ | 284 | | | | | | | | 285 +----+ +----+ +----+ +----+ 286 | | | | | | | | 287 | |-----| | | |-----| | Tier-2 288 | | | | | | | | 289 +----+ +----+ +----+ +----+ 290 | | | | 291 | | | | 292 | +-----+ | | +-----+ | 293 +-| |-+ +-| |-+ Tier-3 294 +-----+ +-----+ 295 | | | | | | 296 <- Servers -> <- Servers -> 298 Figure 1: Typical DC network topology 300 3.2. Clos Network topology 302 This section describes a common design for horizontally scalable 303 topology in large scale data centers in order to meet REQ1. 305 3.2.1. Overview 307 A common choice for a horizontally scalable topology is a folded Clos 308 topology, sometimes called "fat-tree" (see, for example, [INTERCON] 309 and [ALFARES2008]). This topology features an odd number of stages 310 (sometimes known as dimensions) and is commonly made of uniform 311 elements, e.g. network switches with the same port count. Therefore, 312 the choice of folded Clos topology satisfies REQ1 and facilitates 313 REQ2. See Figure 2 below for an example of a folded 3-stage Clos 314 topology (3 stages counting Tier-2 stage twice, when tracing a packet 315 flow): 317 +-------+ 318 | |----------------------------+ 319 | |------------------+ | 320 | |--------+ | | 321 +-------+ | | | 322 +-------+ | | | 323 | |--------+---------+-------+ | 324 | |--------+-------+ | | | 325 | |------+ | | | | | 326 +-------+ | | | | | | 327 +-------+ | | | | | | 328 | |------+-+-------+-+-----+ | | 329 | |------+-+-----+ | | | | | 330 | |----+ | | | | | | | | 331 +-------+ | | | | | | ---------> M links 332 Tier-1 | | | | | | | | | 333 +-------+ +-------+ +-------+ 334 | | | | | | 335 | | | | | | Tier-2 336 | | | | | | 337 +-------+ +-------+ +-------+ 338 | | | | | | | | | 339 | | | | | | ---------> N Links 340 | | | | | | | | | 341 O O O O O O O O O Servers 343 Figure 2: 3-Stage Folded Clos topology 345 This topology is often also referred to as a "Leaf and Spine" 346 network, where "Spine" is the name given to the middle stage of the 347 Clos topology (Tier-1) and "Leaf" is the name of input/output stage 348 (Tier-2). For uniformity, this document will refer to these layers 349 using the "Tier-n" notation. 351 3.2.2. Clos Topology Properties 353 The following are some key properties of the Clos topology: 355 o The topology is fully non-blocking (or more accurately: non- 356 interfering) if M >= N and oversubscribed by a factor of N/M 357 otherwise. Here M and N is the uplink and downlink port count 358 respectively, for a Tier-2 switch as shown in Figure 2. 360 o Utilizing this topology requires control and data plane supporting 361 ECMP with the fan-out of M or more. 363 o Tier-1 switches have exactly one path to every server in this 364 topology. This is an important property that makes route 365 summarization impossible in this topology (see Section 8.2 below). 367 o Traffic flowing from server to server is load-balanced over all 368 available paths using ECMP. 370 3.2.3. Scaling the Clos topology 372 A Clos topology can be scaled either by increasing network element 373 port density or adding more stages, e.g. moving to a 5-stage Clos, as 374 illustrated in Figure 3 below: 376 Tier-1 377 +-----+ 378 | | 379 +--| |--+ 380 | +-----+ | 381 Tier-2 | | Tier-2 382 +-----+ | +-----+ | +-----+ 383 +-------------| DEV |--+--| |--+--| |-------------+ 384 | +-----| C |--+ | | +--| |-----+ | 385 | | +-----+ +-----+ +-----+ | | 386 | | | | 387 | | +-----+ +-----+ +-----+ | | 388 | +-----+-----| DEV |--+ | | +--| |-----+-----+ | 389 | | | +---| D |--+--| |--+--| |---+ | | | 390 | | | | +-----+ | +-----+ | +-----+ | | | | 391 | | | | | | | | | | 392 +-----+ +-----+ | +-----+ | +-----+ +-----+ 393 | DEV | | DEV | +--| |--+ | | | | 394 | A | | B | Tier-3 | | Tier-3 | | | | 395 +-----+ +-----+ +-----+ +-----+ +-----+ 396 | | | | | | | | 397 O O O O O O O O 398 Servers Servers 400 Figure 3: 5-Stage Clos topology 402 The small example topology on Figure 3 is built from devices with a 403 port count of 4 and provides full bisectional bandwidth to all 404 connected servers. In this document, one set of directly connected 405 Tier-2 and Tier-3 devices along with their attached servers will be 406 referred to as a "cluster". For example, DEV A, B, C, D, and the 407 servers that connect to DEV A and B, on Figure 3 form a cluster. 409 In practice, the Tier-3 layer of the network, which are typically top 410 of rack switches (ToRs), is where oversubscription is introduced to 411 allow for packaging of more servers in the data center while meeting 412 the bandwidth requirements for different types of applications. The 413 main reason to limit oversubscription at a single layer of the 414 network is to simplify application development that would otherwise 415 need to account for multiple bandwidth pools: within rack (Tier-3), 416 between racks (Tier-2), and between clusters (Tier-1). Since 417 oversubscription does not have a direct relationship to the routing 418 design it is not discussed further in this document. 420 3.2.4. Managing the Size of Clos Topology Tiers 422 If a data-center network size is small, it is possible to reduce the 423 number of switches in Tier-1 or Tier-2 of Clos topology by a power of 424 two. To understand how this could be done, take Tier-1 as an 425 example. Every Tier-2 device connects to a single group of Tier-1 426 devices. If half of the ports on each of the Tier-1 devices are not 427 being used then it is possible to reduce the number of Tier-1 devices 428 by half and simply map two uplinks from a Tier-2 device to the same 429 Tier-1 device that were previously mapped to different Tier-1 430 devices. This technique maintains the same bisectional bandwidth 431 while reducing the number of elements in the Tier-1 layer, thus 432 saving on CAPEX. The tradeoff, in this example, is the reduction of 433 maximum DC size in terms of overall server count by half. 435 In this example, Tier-2 devices will be using two parallel links to 436 connect to each Tier-1 device. If one of these links fails, the 437 other will pick up all traffic of the failed link, possible resulting 438 in heavy congestion and quality of service degradation if the path 439 determination procedure, does not take bandwidth amount into account. 440 To avoid this situation, parallel links can be grouped in link 441 aggregation groups (LAGs, such as [IEEE8023AD]) with widely available 442 implementation settings that take the whole "bundle" down upon a 443 single link failure. Equivalent techniques that enforce "fate 444 sharing" on the parallel links can be used in place of LAGs to 445 achieve the same effect. As a result of such fate-sharing, traffic 446 from two or more failed links will be re-balanced over the multitude 447 of remaining paths that equals the number of Tier-1 devices. This 448 example is using two links for simplicity it should be noted, that 449 having more links in a bundle will have less impact on capacity upon 450 a member-link failure. 452 4. Data Center Routing Overview 454 This section provides an overview of three general types of data 455 center protocol designs - Layer 2 only, Hybrid L2/L3 and Layer 3 456 only. 458 4.1. Layer 2 Only Designs 460 Originally most data center designs used Spanning-Tree Protocol (STP) 461 for loop free topology creation, typically utilizing variants of the 462 traditional DC topology described in Section 3.1. At the time, many 463 DC switches either did not support Layer 3 routed protocols or 464 supported it with additional licensing fees, which played a part in 465 the design choice. Although many enhancements have been made through 466 the introduction of Rapid Spanning Tree Protocol and Multiple 467 Spanning Tree Protocol that increase convergence, stability and load 468 balancing in larger topologies many of the fundamentals of the 469 protocol limit its applicability in large scale DC's. STP and its 470 newer variants use an active/standby approach to path selection and 471 are therefore hard to deploy in horizontally scaled topologies 472 described in Section 3.2. Further, operators have had many 473 experiences with large failures due to issues caused by improper 474 cabling, misconfiguration, or flawed software on a single device. 475 These failures regularly affected the entire spanning-tree domain and 476 were very hard to troubleshoot due to the nature of the protocol. 477 For these reasons, and since almost all DC traffic is now IP, 478 therefore requiring a Layer 3 routing protocol at the network edge 479 for external connectivity, designs utilizing STP usually fail all of 480 the requirements of large scale DC operators. Various enhancements 481 to link-aggregation protocols such as [IEEE8023AD], generally known 482 as Multi-Chassis Link-Aggregation (M-LAG) made it possible to use 483 Layer 2 designs with active-active network paths while relying on STP 484 as the backup for loop prevention. The major downside of this 485 approach is proprietary nature of such extensions. 487 It should be noted that building large, horizontally scalable, Layer 488 2 only networks without STP is possible recently through the 489 introduction of TRILL [RFC6325]. TRILL resolves many of the issues 490 STP has for large scale DC design however currently the maturity of 491 the protocol, limited number of implementations, and requirement for 492 new equipment that supports it has limited its applicability and 493 increased the cost of such designs. 495 Finally, neither TRILL nor M-LAG approach eliminate the fundamental 496 problem of the shared broadcast domain, that is so detrimental to the 497 operations of any Layer 2, Ethernet based solutions. 499 4.2. Hybrid L2/L3 Designs 501 Operators have sought to limit the impact of data-plane faults and 502 build larger scale topologies through implementing routing protocols 503 in either the Tier-1 or Tier-2 parts of the network and dividing the 504 Layer-2 domain into numerous, smaller domains. This design has 505 allowed data centers to scale up, but at the cost of complexity in 506 the network managing multiple protocols. For the following reasons, 507 operators have retained Layer 2 in either the access (Tier-3) or both 508 access and aggregation (Tier-3 and Tier-2) parts of the network: 510 o Supporting legacy applications that may require direct Layer 2 511 adjacency or use non-IP protocols. 513 o Seamless mobility for virtual machines that require the 514 preservation of IP addresses when a virtual machine moves to 515 different Tier-3 switch. 517 o Simplified IP addressing = less IP subnets is required for the 518 data center. 520 o Application load-balancing may require direct Layer 2 reachability 521 to perform certain functions such as Layer 2 Direct Server Return 522 (DSR). 524 o Continued CAPEX differences between Layer-2 and Layer-3 capable 525 switches. 527 4.3. Layer 3 Only Designs 529 Network designs that leverage IP routing down to Tier-3 of the 530 network have gained popularity as well. The main benefit of these 531 designs is improved network stability and scalability, as a result of 532 confining L2 broadcast domains. Commonly an IGP such as OSPF 533 [RFC2328] is used as the primary routing protocol in such a design. 534 As data centers grow in scale, and server count exceeds tens of 535 thousands, such fully routed designs have become more attractive. 537 Choosing a Layer 3 only design greatly simplifies the network, 538 facilitating the meeting of REQ1 and REQ2, and has widespread 539 adoption in networks where large Layer 2 adjacency and larger size 540 Layer 3 subnets are not as critical compared to network scalability 541 and stability. Application providers and network operators continue 542 to also develop new solutions to meet some of the requirements that 543 previously have driven large Layer 2 domains. 545 5. Routing Protocol Selection and Design 547 In this section the motivations for using External BGP (EBGP) as the 548 single routing protocol for data center networks having a Layer 3 549 protocol design and Clos topology are reviewed. Then, a practical 550 approach for designing an EBGP based network is provided. 552 5.1. Choosing EBGP as the Routing Protocol 554 REQ2 would give preference to the selection of a single routing 555 protocol to reduce complexity and interdependencies. While it is 556 common to rely on an IGP in this situation, sometimes with either the 557 addition of EBGP at the device bordering the WAN or Internal BGP 558 (IBGP) throughout, this document proposes the use of an EBGP only 559 design. 561 Although EBGP is the protocol used for almost all inter-provider 562 routing on the Internet and has wide support from both vendor and 563 service provider communities, it is not generally deployed as the 564 primary routing protocol within the data center for a number of 565 reasons (some of which are interrelated): 567 o BGP is perceived as a "WAN only protocol only" and not often 568 considered for enterprise or data center applications. 570 o BGP is believed to have a "much slower" routing convergence 571 compared to IGPs. 573 o BGP deployment within an Autonomous System typically assumes the 574 presence of an IGP for next-hop resolution. 576 o BGP is perceived to require significant configuration overhead and 577 does not support neighbor auto-discovery. 579 This document discusses some of these perceptions, especially as 580 applicable to the proposed design, and highlights some of the 581 advantages of using the protocol such as: 583 o BGP has less complexity within its protocol design - internal data 584 structures and state-machines are simpler when compared to a link- 585 state IGP such as OSPF. For example, instead of implementing 586 adjacency formation, adjacency maintenance and/or flow-control, 587 BGP simply relies on TCP as the underlying transport. This 588 fulfills REQ2 and REQ3. 590 o BGP information flooding overhead is less when compared to link- 591 state IGPs. Since every BGP router calculates and propagates only 592 the best-path selected, a network failure is masked as soon as the 593 BGP speaker finds an alternate path, which exists when highly 594 symmetric topologies, such as Clos, are coupled with EBGP only 595 design. In contrast, the event propagation scope of a link-state 596 IGP is an entire area, regardless of the failure type. This meets 597 REQ3 and REQ4. It is worth mentioning that all widely deployed 598 link-state IGPs also feature periodic refreshes of routing 599 information, while BGP does not expire routing state, even if this 600 rarely causes significant impact to modern router control planes. 602 o BGP supports third-party (recursively resolved) next-hops. This 603 allows for manipulating multi-path to be non-ECMP based or 604 forwarding based on application-defined forwarding paths, through 605 establishment of a peering session with an application 606 "controller" which can inject routing information into the system, 607 satisfying REQ5. OSPF provides similar functionality using 608 concepts such as "Forwarding Address", but with more difficulty in 609 implementation and lack of protocol simplicity. 611 o Using a well-defined BGP ASN allocation scheme and standard 612 AS_PATH loop detection, "BGP path hunting" (see [JAKMA2008]) can 613 be controlled and complex unwanted paths will be ignored. See 614 Section 5.2 for an example of a working BGP ASN allocation scheme. 615 In a link-state IGP accomplishing the same goal would require 616 multi-(instance/topology/processes) support, typically not 617 available in all DC devices and quite complex to configure and 618 troubleshoot. Using a traditional single flooding domain, which 619 most DC designs utilize, under certain failure conditions may pick 620 up unwanted lengthy paths, e.g. traversing multiple Tier-2 621 devices. 623 o EBGP configuration that is implemented with minimal routing policy 624 is easier to troubleshoot for network reachability issues. In 625 most implementations, it is straightforward to view contents of 626 BGP Loc-RIB and compare it to the router's RIB. Also every BGP 627 neighbor has corresponding Adj-RIB-In and Adj-RIB-Out structures 628 with incoming and outgoing NRLI information that can be easily 629 correlated on both sides of a BGP session. Thus, BGP satisfies 630 REQ3. 632 5.2. EBGP Configuration for Clos topology 634 Clos topologies that have more than 5 stages are very uncommon due to 635 the large numbers of interconnects required by such a design. 636 Therefore, the examples below are made with reference to the 5-stage 637 Clos topology (5 stages in unfolded state). 639 5.2.1. Example ASN Scheme 641 The diagram below illustrates an example ASN allocation scheme. The 642 following is a list of guidelines that can be used: 644 o Only EBGP sessions established over direct point-to-point links 645 interconnecting the network nodes. 647 o 16-bit (two octet) BGP ASNs are used, since these are widely 648 supported and have better vendor interoperability. 650 o Private BGP ASNs from the range 64512-65534 are used so as to 651 avoid ASN conflicts. 653 o A single BGP ASN is allocated to all of the Clos topology's Tier-1 654 devices. 656 o Unique BGP ASN is allocated per each group of Tier-2 devices. 658 o Unique BGP ASN is allocated to every Tier-3 device (e.g. ToR) in 659 this topology. 661 ASN 65534 662 +---------+ 663 | +-----+ | 664 | | | | 665 +-|-| |-|-+ 666 | | +-----+ | | 667 ASN 646XX | | | | ASN 646XX 668 +---------+ | | | | +---------+ 669 | +-----+ | | | +-----+ | | | +-----+ | 670 +-----------|-| |-|-+-|-| |-|-+-|-| |-|-----------+ 671 | +---|-| |-|-+ | | | | +-|-| |-|---+ | 672 | | | +-----+ | | +-----+ | | +-----+ | | | 673 | | | | | | | | | | 674 | | | | | | | | | | 675 | | | +-----+ | | +-----+ | | +-----+ | | | 676 | +-----+---|-| |-|-+ | | | | +-|-| |-|---+-----+ | 677 | | | +-|-| |-|-+-|-| |-|-+-|-| |-|-+ | | | 678 | | | | | +-----+ | | | +-----+ | | | +-----+ | | | | | 679 | | | | +---------+ | | | | +---------+ | | | | 680 | | | | | | | | | | | | 681 +-----+ +-----+ | | +-----+ | | +-----+ +-----+ 682 | ASN | | | +-|-| |-|-+ | | | | 683 |65YYY| | ... | | | | | | ... | | ... | 684 +-----+ +-----+ | +-----+ | +-----+ +-----+ 685 | | | | +---------+ | | | | 686 O O O O <- Servers -> O O O O 688 Figure 4: BGP ASN layout for 5-stage Clos 690 5.2.2. Private Use BGP ASNs 692 The original range of Private Use BGP ASNs [RFC6996] limited 693 operators to 1023 unique ASNs. Since it is quite likely that the 694 number of network devices may exceed this number, a workaround is 695 required. One approach is to re-use the ASNs assigned to the Tier-3 696 devices across different clusters. For example, Private Use BGP ASNs 697 65001, 65002 ... 65032 could be used within every individual cluster 698 and assigned to Tier-3 devices. 700 To avoid route suppression due to the AS_PATH loop detection 701 mechanism in BGP, upstream EBGP sessions on Tier-3 devices must be 702 configured with the "AllowAS In" feature that allows accepting a 703 device's own ASN in received route advertisements. Introducing this 704 feature does not create an opportunity for routing loops under 705 misconfiguration since the AS_PATH is always incremented when routes 706 are propagated between topology tiers. Loop protection is also in 707 place at the Tier-1 device, which does not accept routes with a path 708 including its own ASN. 710 Another solution to this problem would be using four-octet BGP ASNs 711 ([RFC6793]), where there are additional Private Use ASN's available, 712 see [IANA.AS]. Use of Four-Octet BGP ASNs put additional protocol 713 complexity in the BGP implementation so should be considered against 714 the complexity of re-use when considering REQ3 and REQ4. Perhaps 715 more importantly, they are not yet supported by all BGP 716 implementations, which may limit vendor selection of DC equipment. 718 5.2.3. Prefix Advertisement 720 A Clos topology features a large number of point-to-point links and 721 associated prefixes. Advertising all of these routes into BGP may 722 create FIB overload conditions in the network devices. Advertising 723 these links also puts additional path computation stress on the BGP 724 control plane for little benefit. There are two possible solutions: 726 o Do not advertise any of the point-to-point links into BGP. Since 727 the EBGP based design changes the next-hop address at every 728 device, distant networks will automatically be reachable via the 729 advertising EBGP peer and do not require reachability to these 730 prefixes. However, this may complicate operational 731 troubleshooting or monitoring systems if the addresses are not 732 reachable: e.g. using the popular "traceroute" tool will display 733 IP addresses that are not reachable. 735 o Advertise point-to-point links, but summarize them on every 736 device. This requires an address allocation scheme such as 737 allocating a consecutive block of IP addresses per Tier-1 and 738 Tier-2 device to be used for point-to-point interface addressing 739 to the lower layers (Tier-2 uplinks will be numbered out of Tier-1 740 addressing and so forth). 742 Server subnets on Tier-3 devices must be announced into BGP without 743 using route summarization on Tier-2 and Tier-1 devices. Summarizing 744 subnets in a Clos topology results in route black-holing under a 745 single link failure (e.g. between Tier-2 and Tier-3 devices) and 746 hence must be avoided. The use of peer links within the same tier to 747 resolve the black-holing problem by providing "bypass paths" is 748 undesirable due to O(N^2) complexity of the peering mesh and waste of 749 ports on the devices. An alternative to the full-mesh of peer-links 750 would be using a simpler bypass topology, e.g. a "ring" as described 751 in [FB4POST], but such a topology adds extra hops and has very 752 limited bisection bandwidth, in addition requiring special tweaks to 753 make BGP routing work - such as possibly splitting every device into 754 an ASN on its own. In Section 8.2 another, less intrusive, method 755 for performing a limited form route summarization in Clos networks 756 and the associated trade-offs are described. 758 5.2.4. External Connectivity 760 A dedicated cluster (or clusters) in the Clos topology could be used 761 for the purpose of connecting to the Wide Area Network (WAN) edge 762 devices, or WAN Routers. Tier-3 devices in such cluster would be 763 replaced with WAN routers, and EBGP peering would be used again, 764 though WAN routers are likely to belong to a public ASN if Internet 765 connectivity is required in the design. The Tier-2 devices in such a 766 dedicated cluster will be referred to as "Border Routers" in this 767 document. These devices have to perform a few special functions: 769 o Hide network topology information when advertising paths to WAN 770 routers, i.e. remove Private BGP ASNs from the AS_PATH attribute. 771 This is typically done to avoid ASN number collisions between 772 different data centers. An implementation specific BGP feature 773 typically called "Remove Private AS" is commonly used to 774 accomplish this. Depending on implementation, the feature should 775 strip a contiguous sequence of private ASNs found in AS_PATH 776 attribute prior to advertising the path to a neighbor. This 777 assumes that all BGP ASN's used for intra data center numbering 778 are from the private ASN range. The process for stripping the 779 private ASNs is not currently standardized, but most 780 implementations commonly follow the logic described in 781 [REMOVE-PRIVATE-AS] vendor's document. 783 o Originate a default route to the data center devices. This is the 784 only place where default route can be originated, as route 785 summarization is risky for the "scale-out" topology. 786 Alternatively, Border Routers may simply relay the default route 787 learned from WAN routers. Advertising the default route from 788 Border Routers requires that all Border Routers to be fully 789 connected to the WAN Routers upstream, to provide resistance to a 790 single-link failure causing the black holing of traffic. To 791 prevent chance of operator or implementation error that may impact 792 EBGP sessions to the WAN routers simultaneously (although these 793 scenarios are not planned for by many operators since they 794 represents a multiple failure) it is more desirable to take this 795 approach rather than introducing complicated conditional default 796 origination schemes provided by some implementations. 798 5.2.5. Route Summarization at the Edge 800 It is often desirable to summarize network reachability information 801 prior to advertising it to the WAN network due to high amount of IP 802 prefixes originated from within the data center in a fully routed 803 network design. For example, a network with 2000 Tier-3 devices will 804 have at least 2000 servers subnets advertised into BGP, along with 805 the infrastructure or other prefixes. However, as discussed before, 806 the proposed network design does not allow for route summarization 807 due to the lack of peer links inside every tier. 809 However, it is possible to lift this restriction for the Border 810 Routers, by devising a different connectivity model for these 811 devices. There are two options possible: 813 o Interconnect the Border Routers using a full-mesh of physical 814 links or using any other "peer-mesh" topology, such as ring or 815 hub-and-spoke. Configure BGP accordingly on all Border Leafs to 816 exchange network reachability information - e.g. by adding a mesh 817 of iBGP sessions. The interconnecting peer links need to be 818 appropriately sized for traffic that will be present in the case 819 of a device or link failure underneath the Border Routers. 821 o Tier-1 devices may have additional physical links provisioned 822 toward the Border Routers (which are Tier-2 devices from the 823 perspective of Tier-1). Specifically, if protection from a single 824 link or node failure is desired, each Tier-1 devices would have to 825 connect to at least two Border Routers. This puts additional 826 requirements on the port count for Tier-1 devices and Border 827 Routers, potentially making it a non-uniform, larger port count, 828 device with the other devices in the Clos. This also reduces the 829 number of ports available to "regular" Tier-2 switches and hence 830 the number of clusters that could be interconnected via Tier-1 831 layer. 833 If any of the above options are implemented, it is possible to 834 perform route summarization at the Border Routers toward the WAN 835 network core without risking a routing black-hole condition under a 836 single link failure. Both of the options would result in non-uniform 837 topology as additional links have to be provisioned on some network 838 devices. 840 6. ECMP Considerations 842 This section covers the Equal Cost Multipath (ECMP) functionality for 843 Clos topology and discusses a few special requirements. 845 6.1. Basic ECMP 847 ECMP is the fundamental load-sharing mechanism used by a Clos 848 topology. Effectively, every lower-tier device will use all of its 849 directly attached upper-tier devices to load-share traffic destined 850 to the same IP prefix. Number of ECMP paths between any two Tier-3 851 devices in Clos topology equals to the number of the devices in the 852 middle stage (Tier-1). For example, Figure 5 illustrates the 853 topology where Tier-3 device A has four paths to reach servers X and 854 Y, via Tier-2 devices B and C and then Tier-1 devices 1, 2, 3, and 4 855 respectively. 857 Tier-1 858 +-----+ 859 | DEV | 860 +->| 1 |--+ 861 | +-----+ | 862 Tier-2 | | Tier-2 863 +-----+ | +-----+ | +-----+ 864 +------------>| DEV |--+->| DEV |--+--| |-------------+ 865 | +-----| B |--+ | 2 | +--| |-----+ | 866 | | +-----+ +-----+ +-----+ | | 867 | | | | 868 | | +-----+ +-----+ +-----+ | | 869 | +-----+---->| DEV |--+ | DEV | +--| |-----+-----+ | 870 | | | +---| C |--+->| 3 |--+--| |---+ | | | 871 | | | | +-----+ | +-----+ | +-----+ | | | | 872 | | | | | | | | | | 873 +-----+ +-----+ | +-----+ | +-----+ +-----+ 874 | DEV | | | Tier-3 +->| DEV |--+ Tier-3 | | | | 875 | A | | | | 4 | | | | | 876 +-----+ +-----+ +-----+ +-----+ +-----+ 877 | | | | | | | | 878 O O O O <- Servers -> X Y O O 880 Figure 5: ECMP fan-out tree from A to X and Y 882 The ECMP requirement implies that the BGP implementation must support 883 multi-path fan-out for up to the maximum number of devices directly 884 attached at any point in the topology in upstream or downstream 885 direction. Normally, this number does not exceed half of the ports 886 found on a device in the topology. For example, an ECMP fan-out of 887 32 would be required when building a Clos network using 64-port 888 devices. The Border Routers may need to have wider fan-out to be 889 able to connect to multitude of Tier-1 devices if route summarization 890 at Border Router level is implemented as described in Section 5.2.5. 891 If a device's hardware does not support wider ECMP, logical link- 892 grouping (link-aggregation at layer 2) could be used to provide 893 "hierarchical" ECMP (Layer 3 ECMP followed by Layer 2 ECMP) to 894 compensate for fan-out limitations. Such approach, however, 895 increases the risk of flow polarization, as less entropy will be 896 available to the second stage of ECMP. 898 Most BGP implementations declare paths to be equal from ECMP 899 perspective if they match up to and including step (e) 900 Section 9.1.2.2 of [RFC4271]. In the proposed network design there 901 is no underlying IGP, so all IGP costs are assumed to be zero or 902 otherwise the same value across all paths and policies may be applied 903 as necessary to equalize BGP attributes that vary in vendor defaults, 904 as has been seen occasionally with MED and origin code. Routing 905 loops are unlikely due to the BGP best-path selection process which 906 prefers shorter AS_PATH length, and longer paths through the Tier-1 907 devices which don't allow their own AS in the path and have the same 908 ASN are also not possible. 910 6.2. BGP ECMP over Multiple ASNs 912 For application load-balancing purposes it is desirable to have the 913 same prefix advertised from multiple Tier-3 devices. From the 914 perspective of other devices, such a prefix would have BGP paths with 915 different AS_PATH attribute values, while having the same AS_PATH 916 attribute lengths. Therefore, BGP implementations must support load- 917 sharing over above-mentioned paths. This feature is sometimes known 918 as "multipath relax" and effectively allows for ECMP to be done 919 across different neighboring ASNs if all other attributes are equal 920 as described in the previous section. 922 6.3. Weighted ECMP 924 It may be desirable for the network devices to implement weighted 925 ECMP, to be able to send more traffic over some paths in ECMP fan- 926 out. This could be helpful to compensate for failures in the network 927 and send more traffic over paths that have more capacity. The 928 prefixes that require weighted ECMP would have to be injected using 929 remote BGP speaker (central agent) over a multihop session as 930 described further in Section 8.1. If support in implementations is 931 available, weight-distribution for multiple BGP paths could be 932 signaled using the technique described in 933 [I-D.ietf-idr-link-bandwidth]. 935 6.4. Consistent Hashing 937 It is often desirable to have the hashing function used to ECMP to be 938 consistent (see [CONS-HASH]), to minimizing the impact on flow to 939 next-hop affinity changes when a next-hop is added or removed to ECMP 940 group. This could be used if the network device is used as a load- 941 balancer, mapping flows toward multiple destinations - in this case, 942 losing or adding a destination will not have detrimental effect of 943 currently established flows. One particular recommendation on 944 implementing consistent hashing is provided in [RFC2992], though 945 other implementations are possible. This functionality could be 946 naturally combined with weighted ECMP, with the impact of the next- 947 hop changes being proportional to the weight of the given next-hop. 948 Notice that the usual downside of consistent hashing is increased 949 load on hardware resource utilization, as typically more space is 950 required to implement a consistent-hashing region. 952 7. Routing Convergence Properties 954 This section reviews routing convergence properties in the proposed 955 design. A case is made that sub-second convergence is achievable if 956 the implementation supports fast EBGP peering session deactivation 957 and timely RIB and FIB update upon failure of the associated link. 959 7.1. Fault Detection Timing 961 BGP typically relies on an IGP to route around link/node failures 962 inside an AS, and implements either a polling based or an event- 963 driven mechanism to obtain updates on IGP state changes. The 964 proposed routing design does not use an IGP, so the only mechanisms 965 that could be used for fault detection are BGP keep-alive process (or 966 any other type of keep-alive mechanism) and link-failure triggers. 968 Relying solely on BGP keep-alive packets may result in high 969 convergence delays, in the order of multiple seconds (on many BGP 970 implementations the minimum configurable BGP hold timer value is 971 three seconds). However, many BGP implementations can shut down 972 local EBGP peering sessions in response to the "link down" event for 973 the outgoing interface used for BGP peering. This feature is 974 sometimes called as "fast fallover". Since links in modern data 975 centers are often point-to-point fiber connections, a physical 976 interface failure is often detected in milliseconds and subsequently 977 triggers a BGP re-convergence. 979 Ethernet technologies may support failure signaling or detection 980 standards such as [IEEE8021AG] and [IEEE8023AH], which may make 981 failure detection more robust. Alternatively, some platforms may 982 support Bidirectional Forwarding Detection (BFD) [RFC5880] to allow 983 for sub-second failure detection and fault signaling to the BGP 984 process. However, use of either of these presents additional 985 requirements to vendor software and possibly hardware, and may 986 contradict REQ1. Until recently with [RFC7130], BFD also did not 987 allow detection of a single member link failure on a LAG, which would 988 limit's it's usefulness in some designs. 990 7.2. Event Propagation Timing 992 In this design the impact of BGP Minimum Route Advertisement Interval 993 (MRAI) timer (See section 9.2.1.1 of [RFC4271]) should be considered. 994 Per the standard it is required for BGP implementations to space out 995 consecutive BGP UPDATE messages by at least MRAI seconds, which is 996 often a configurable value. The initial BGP UPDATE messages after an 997 event carrying withdrawn routes are commonly not affected by this 998 timer. The MRAI timer may present significant convergence delays 999 when a BGP speaker "waits" for the new path to be learned from its 1000 peers and has no local backup path information. 1002 In a Clos topology each EBGP speaker has either one path or N paths 1003 for the same prefix, where N is a significantly large number, e.g. 1004 N=32 (the ECMP fan-out). Therefore, if a path fails there is either 1005 no backup path at all, or the backup is readily available in BGP Loc- 1006 RIB. In the former case, the BGP withdrawal announcement will 1007 propagate un-delayed and trigger re-convergence on affected devices. 1008 In the latter case, the best-path will be re-evaluated and the local 1009 ECMP group corresponding to the new next-hop set changed. If the BGP 1010 path was the best-path selected previously, an "implicit withdraw" 1011 will be sent via a BGP UPDATE message as described as option b in 1012 Section 3.1 of [RFC4271] due to the BGP AS_PATH attribute changing. 1014 7.3. Impact of Clos Topology Fan-outs 1016 Clos topology has large fan-outs, which may impact the "Up->Down" 1017 convergence in some cases, as described in this section. In a 1018 situation when a link between Tier-3 and Tier-2 device fails, the 1019 Tier-2 device will send BGP WITHDRAW message to all upstream Tier-1 1020 devices, and Tier-1 devices will relay this message to all downstream 1021 Tier-2 devices (except for the originator). Tier-2 devices other 1022 than the one originating the WITHDRAW should then wait for ALL 1023 adjacent Tier-1 devices to send a WITHDRAW message before it removes 1024 the affected prefixes and sends corresponding WITHDRAW downstream to 1025 connected Tier-3 devices. If the original Tier-2 device or the 1026 relaying Tier-1 devices introduce some delay into their 1027 announcements, the result could be WITHDRAW message "dispersion", 1028 that could be as long as multiple seconds. In order to avoid such 1029 behavior, BGP implementations must support "update groups". The 1030 "update group" is defined as a collection of neighbors sharing the 1031 same outbound policy - the local speaker will send BGP updates to the 1032 members of the group synchronously. 1034 The impact of such "dispersion" grows with the size of topology fan- 1035 out and could also grow under network convergence churn. 1037 7.4. Failure Impact Scope 1039 A network is declared to converge in response to a failure once all 1040 devices within the failure impact scope are notified of the event and 1041 have re-calculated their RIB's and consequently updated their FIB's. 1042 Larger failure impact scope typically means slower convergence since 1043 more devices have to be notified, and additionally results in a less 1044 stable network. In this section we describe BGP's advantages over 1045 link-state routing protocols in reducing failure impact scope for a 1046 Clos topology. 1048 BGP is behaves like a distance-vector protocol in the sense that only 1049 the best path from the point of view of the local router is sent to 1050 neighbors. As such, some failures are masked if the local node can 1051 immediately find a backup path and does not have to send any updates 1052 further. Notice that in the worst case ALL devices in a data center 1053 topology have to either withdraw a prefix completely or update the 1054 ECMP groups in the FIB. However, many failures will not result in 1055 such a wide impact. There are two main failure types where impact 1056 scope is reduced: 1058 o Failure of a link between Tier-2 and Tier-1 devices: In this case, 1059 a Tier-2 device will update the affected ECMP groups, removing the 1060 failed link. There is no need to send new information to 1061 downstream Tier-3 devices, unless the path was selected as best by 1062 the BGP process, in which case only an "implicit withdraw" needs 1063 to be sent, which should not affect forwarding. The affected 1064 Tier-1 device will lose the only path available to reach a 1065 particular cluster and will have to withdraw the associated 1066 prefixes. Such prefix withdrawal process will only affect Tier-2 1067 devices directly connected to the affected Tier-1 device. The 1068 Tier-2 devices receiving the BGP UPDATE messages withdrawing 1069 prefixes will simply have to update their ECMP groups. The Tier-3 1070 devices are not involved in the re-convergence process. 1072 o Failure of a Tier-1 device: In this case, all Tier-2 devices 1073 directly attached to the failed node will have to update their 1074 ECMP groups for all IP prefixes from non-local cluster. The 1075 Tier-3 devices are once again not involved in the re-convergence 1076 process, but may receive "implicit withdraws" as described above. 1078 Even though in case of such failures multiple IP prefixes will have 1079 to be reprogrammed in the FIB, it is worth noting that ALL of these 1080 prefixes share a single ECMP group on Tier-2 device. Therefore, in 1081 the case of implementations with a hierarchical FIB, only a single 1082 change has to be made to the FIB. Hierarchical FIB here means FIB 1083 structure where the next-hop forwarding information is stored 1084 separately from the prefix lookup table, and the latter only store 1085 pointers to the respective forwarding information. 1087 Even though BGP offers some failure scope reduction, reduction of the 1088 fault domain using summarization is not always possible with the 1089 proposed design, since using this technique may create routing black- 1090 holes as mentioned previously. Therefore, the worst control-plane 1091 failure impact scope is the network as a whole, for instance in a 1092 case of a link failure between Tier-2 and Tier-3 devices. The amount 1093 of impacted prefixes in this case would be much less than in the case 1094 of a failure in the upper layers of a Clos network topology. The 1095 property of having such large failure scope is not a result of 1096 choosing EBGP in the design but rather a result of using the "scale- 1097 out" Clos topology. 1099 7.5. Routing Micro-Loops 1101 When a downstream device, e.g. Tier-2 device, loses all paths for a 1102 prefix, it normally has the default route pointing toward the 1103 upstream device, in this case the Tier-1 device. As a result, it is 1104 possible to get in the situation when Tier-2 switch loses a prefix, 1105 but Tier-1 switch still has the path pointing to the Tier-2 device, 1106 which results in transient micro-loop, since Tier-1 switch will keep 1107 passing packets to the affected prefix back to Tier-2 device, and 1108 Tier-2 will bounce it back again using the default route. This 1109 micro-loop will last for the duration of time it takes the upstream 1110 device to fully update its forwarding tables. 1112 To minimize impact of the micro-loops, Tier-2 and Tier-1 switches can 1113 be configured with static "discard" or "null" routes that will be 1114 more specific than the default route for specific prefixes missing 1115 during network convergence. For Tier-2 switches, the discard route 1116 should be a summary route, covering all server subnets of the 1117 underlying Tier-3 devices. For Tier-1 devices, the discard route 1118 should be a summary covering the server IP address subnet allocated 1119 for the whole data-center. Those discard routes will only take 1120 precedence for the duration of network convergence, until the device 1121 learns a more specific prefix via a new path. 1123 8. Additional Options for Design 1125 8.1. Third-party Route Injection 1127 BGP allows for a "third-party", i.e. directly attached, BGP speaker 1128 to inject routes anywhere in the network topology, meeting REQ5. 1129 This can be achieved by peering via a multihop BGP session with some 1130 or even all devices in the topology. Furthermore, BGP diverse path 1131 distribution [RFC6774] could be used to inject multiple BGP next hops 1132 for the same prefix to facilitate load-balancing, or using the BGP 1133 ADD-PATH capability [I-D.ietf-idr-add-paths] if supported by the 1134 implementation. Unfortunately, in many implementations ADD-PATH has 1135 been found to only support IBGP properly due to the use cases it was 1136 originally optimized for, which limits the "third-party" peering to 1137 iBGP only, if the feature is used. 1139 To implement route injection in the proposed design a third-party BGP 1140 speaker may peer with Tier-3 and Tier-1 switches, injecting the same 1141 prefix, but using a special set of BGP next-hops for Tier-1 devices. 1142 Those next-hops are assumed to resolve recursively via BGP, and could 1143 be, for example, IP addresses on Tier-3 devices. The resulting 1144 forwarding table programming could provide desired traffic proportion 1145 distribution among different clusters. 1147 8.2. Route Summarization within Clos Topology 1149 As mentioned previously, route summarization is not possible within 1150 the proposed Clos topology since it makes the network susceptible to 1151 route black-holing under single link failures. The main problem is 1152 the limited number of parallel paths between network elements, e.g. 1153 there is only a single path between any pair of Tier-1 and Tier-3 1154 devices. However, some operators may find route aggregation 1155 desirable to improve control plane stability. 1157 Route summarization would be possible with a small modification to 1158 the network topology, though the trade-off would be reduction of the 1159 total size of the network as well as network congestion under 1160 specific failures. This approach is very similar to the technique 1161 described above, which allows Border Routers to summarize the entire 1162 data-center address space. 1164 8.2.1. Collapsing Tier-1 Devices Layer 1166 In order to add more paths between Tier-1 and Tier-3 devices, group 1167 Tier-2 devices into pairs, and then connect the pairs to the same 1168 group of Tier-1 devices. This is logically equivalent to 1169 "collapsing" Tier-1 devices into a group of half the size, merging 1170 the links on the "collapsed" devices. The result is illustrated in 1171 Figure 6. For example, in this topology DEV C and DEV D connect to 1172 the same set of Tier-1 devices (DEV 1 and DEV 2), whereas before they 1173 were connecting to different groups of Tier-1 devices. 1175 Tier-2 Tier-1 Tier-2 1176 +-----+ +-----+ +-----+ 1177 +-------------| DEV |------| DEV |------| |-------------+ 1178 | +-----| C |--++--| 1 |--++--| |-----+ | 1179 | | +-----+ || +-----+ || +-----+ | | 1180 | | || || | | 1181 | | +-----+ || +-----+ || +-----+ | | 1182 | +-----+-----| DEV |--++--| DEV |--++--| |-----+-----+ | 1183 | | | +---| D |------| 2 |------| |---+ | | | 1184 | | | | +-----+ +-----+ +-----+ | | | | 1185 | | | | | | | | 1186 +-----+ +-----+ +-----+ +-----+ 1187 | DEV | | DEV | | | | | 1188 | A | | B | Tier-3 Tier-3 | | | | 1189 +-----+ +-----+ +-----+ +-----+ 1190 | | | | | | | | 1191 O O O O <- Servers -> O O O O 1193 Figure 6: 5-Stage Clos topology 1195 Having this design in place, Tier-2 devices may be configured to 1196 advertise only a default route down to Tier-3 devices. If a link 1197 between Tier-2 and Tier-3 fails, the traffic will be re-routed via 1198 the second available path known to a Tier-2 switch. It is not 1199 possible to advertise a summary route covering prefixes for a single 1200 cluster from Tier-2 devices since each of them has only a single path 1201 down to this prefix. It would require dual-homed servers to 1202 accomplish that. Also note that this design is only resilient to 1203 single link failure. It is possible for a double link failure to 1204 isolate a Tier-2 device from all paths toward a specific Tier-3 1205 device, thus causing a routing black-hole. 1207 A result of the proposed topology modification would be reduction of 1208 Tier-1 devices port capacity. This limits the maximum number of 1209 attached Tier-2 devices and therefore will limit the maximum DC 1210 network size. A larger network would require different Tier-1 1211 devices that have higher port density to implement this change. 1213 Another problem is traffic re-balancing under link failures. Since 1214 three are two paths from Tier-1 to Tier-3, a failure of the link 1215 between Tier-1 and Tier-2 switch would result in all traffic that was 1216 taking the failed link to switch to the remaining path. This will 1217 result in doubling of link utilization on the remaining link. 1219 8.2.2. Simple Virtual Aggregation 1221 A completely different approach to route summarization is possible, 1222 provided that the main goal is to reduce the FIB pressure, while 1223 allowing the control plane to disseminate full routing information. 1224 Firstly, it could be easily noted that in many cases multiple 1225 prefixes, some of which are less specific, share the same set of the 1226 next-hops (same ECMP group). For example, looking from the 1227 perspective of a Tier-3 devices, all routes learned from upstream 1228 Tier-2's, including the default route, will share the same set of BGP 1229 next-hops, provided that there is no failures in the network. This 1230 makes it possible to use the technique similar to described in 1231 [RFC6769] and only install the least specific route in the FIB, 1232 ignoring more specific routes if they share the same next-hop set. 1233 For example, under normal network conditions, only the default route 1234 need to be programmed into FIB. 1236 Furthermore, if the Tier-2 devices are configured with summary 1237 prefixes covering all of their attached Tier-3 device's prefixes the 1238 same logic could be applied in Tier-1 devices as well, and, by 1239 induction to Tier-2/Tier-3 switches in different clusters. These 1240 summary routes should still allow for more specific prefixes to leak 1241 to Tier-1 devices, to enable for detection of mismatches in the next- 1242 hop sets if a particular link fails, changing the next-hop set for a 1243 specific prefix. 1245 Re-stating once again, this technique does not reduce the amount of 1246 control plane state (i.e. BGP UPDATEs/BGP LocRIB sizing), but only 1247 allows for more efficient FIB utilization, by spotting more specific 1248 prefixes that share their next-hops with less specifics. 1250 8.3. ICMP Unreachable Message Masquerading 1252 This section discusses some operational aspects of not advertising 1253 point-to-point link subnets into BGP, as previously outlined as an 1254 option in Section 5.2.3. The operational impact of this decision 1255 could be seen when using the well-known "traceroute" tool. 1256 Specifically, IP addresses displayed by the tool will be the link's 1257 point-to-point addresses, and hence will be unreachable for 1258 management connectivity. This makes some troubleshooting more 1259 complicated. 1261 One way to overcome this limitation is by using the DNS subsystem to 1262 create the "reverse" entries for the IP addresses of the same device 1263 pointing to the same name. The connectivity then can be made by 1264 resolving this name to the "primary" IP address of the devices, e.g. 1265 its Loopback interface, which is always advertised into BGP. 1267 However, this create dependency on DNS subsystem, which may happen to 1268 be unavailable during an outage. 1270 Another option is to make the network device perform IP address 1271 masquerading, that is rewriting the source IP addresses of the 1272 appropriate ICMP messages sent off of the device with the "primary" 1273 IP address of the device. Specifically, the ICMP Destination 1274 Unreachable Message (type 3) codes 3 (port unreachable) and ICMP Time 1275 Exceeded (type 11) code 0, which are involved in proper working of 1276 the "traceroute" tool. With this modification, the "traceroute" 1277 probes sent to the devices will always be sent back with the 1278 "primary" IP address as the source, allowing the operator to discover 1279 the "reachable" IP address of the box. 1281 9. Security Considerations 1283 The design does not introduce any additional security concerns. 1284 General BGP security considerations are discussed in [RFC4271] and 1285 [RFC4272]. Furthermore, the Generalized TTL Security Mechanism 1286 [RFC5082] could be used to reduce the risk of BGP session spoofing. 1288 10. IANA Considerations 1290 This document includes no request to IANA. 1292 11. Acknowledgements 1294 This publication summarizes work of many people who participated in 1295 developing, testing and deploying the proposed network design, some 1296 of whom were George Chen, Parantap Lahiri, Dave Maltz, Edet Nkposong, 1297 Robert Toomey, and Lihua Yuan. Authors would also like to thank 1298 Linda Dunbar, Susan Hares, Russ White and Robert Raszuk for reviewing 1299 the document and providing valuable feedback and Mary Mitchell for 1300 grammar and style suggestions. 1302 12. References 1304 12.1. Normative References 1306 [RFC4271] Rekhter, Y., Li, T., and S. Hares, "A Border Gateway 1307 Protocol 4 (BGP-4)", RFC 4271, January 2006. 1309 [RFC6996] Mitchell, J., "Autonomous System (AS) Reservation for 1310 Private Use", BCP 6, RFC 6996, July 2013. 1312 12.2. Informative References 1314 [RFC2328] Moy, J., "OSPF Version 2", STD 54, RFC 2328, April 1998. 1316 [RFC4272] Murphy, S., "BGP Security Vulnerabilities Analysis", RFC 1317 4272, January 2006. 1319 [RFC4786] Abley, J. and K. Lindqvist, "Operation of Anycast 1320 Services", BCP 126, RFC 4786, December 2006. 1322 [RFC5082] Gill, V., Heasley, J., Meyer, D., Savola, P., and C. 1323 Pignataro, "The Generalized TTL Security Mechanism 1324 (GTSM)", RFC 5082, October 2007. 1326 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 1327 (BFD)", RFC 5880, June 2010. 1329 [RFC6325] Perlman, R., Eastlake, D., Dutt, D., Gai, S., and A. 1330 Ghanwani, "Routing Bridges (RBridges): Base Protocol 1331 Specification", RFC 6325, July 2011. 1333 [RFC6774] Raszuk, R., Fernando, R., Patel, K., McPherson, D., and K. 1334 Kumaki, "Distribution of Diverse BGP Paths", RFC 6774, 1335 November 2012. 1337 [RFC6793] Vohra, Q. and E. Chen, "BGP Support for Four-Octet 1338 Autonomous System (AS) Number Space", RFC 6793, December 1339 2012. 1341 [RFC2992] Hopps, C., "Analysis of an Equal-Cost Multi-Path 1342 Algorithm", RFC 2992, November 2000. 1344 [RFC6769] Raszuk, R., Heitz, J., Lo, A., Zhang, L., and X. Xu, 1345 "Simple Virtual Aggregation (S-VA)", RFC 6769, October 1346 2012. 1348 [RFC7130] Bhatia, M., Chen, M., Boutros, S., Binderberger, M., and 1349 J. Haas, "Bidirectional Forwarding Detection (BFD) on Link 1350 Aggregation Group (LAG) Interfaces", RFC 7130, February 1351 2014. 1353 [I-D.ietf-idr-add-paths] 1354 Walton, D., Retana, A., Chen, E., and J. Scudder, 1355 "Advertisement of Multiple Paths in BGP", draft-ietf-idr- 1356 add-paths-09 (work in progress), October 2013. 1358 [I-D.ietf-idr-link-bandwidth] 1359 Mohapatra, P. and R. Fernando, "BGP Link Bandwidth 1360 Extended Community", draft-ietf-idr-link-bandwidth-06 1361 (work in progress), January 2013. 1363 [GREENBERG2009] 1364 Greenberg, A., Hamilton, J., and D. Maltz, "The Cost of a 1365 Cloud: Research Problems in Data Center Networks", January 1366 2009. 1368 [IEEE8021AG] 1369 IEEE 802.1Q, , "IEEE Standard for Local and metropolitan 1370 area networks - Media Access Control (MAC) Bridges and 1371 Virtual Bridged Local Area Networks", October 2012. 1373 [IEEE8023AH] 1374 IEEE 802.3, , "IEEE Standard for Information technology - 1375 Local and metropolitan area networks - Carrier sense 1376 multiple access with collision detection (CSMA/CD) access 1377 method and physical layer specifications", December 2008. 1379 [INTERCON] 1380 Dally, W. and B. Towles, "Principles and Practices of 1381 Interconnection Networks", ISBN 978-0122007514, January 1382 2004. 1384 [ALFARES2008] 1385 Al-Fares, M., Loukissas, A., and A. Vahdat, "A Scalable, 1386 Commodity Data Center Network Architecture", August 2008. 1388 [IANA.AS] IANA, , "Autonomous System (AS) Numbers", August 2014, 1389 . 1391 [IEEE8023AD] 1392 IEEE 802.3ad, , "IEEE Standard for Link aggregation for 1393 parallel links", October 2000. 1395 [REMOVE-PRIVATE-AS] 1396 Cisco Systems, , "Removing Private Autonomous System 1397 Numbers in BGP", August 2005, 1398 . 1401 [FB4POST] Farrington, N. and A. Andreyev, "Facebook's Data Center 1402 Network Architecture", May 2013, 1403 . 1405 [JAKMA2008] 1406 Jakma, P., "BGP Path Hunting", 2008, 1407 . 1409 [CONS-HASH] 1410 Wikipedia, , "Consistent Hashing", 1411 . 1413 Authors' Addresses 1415 Petr Lapukhov 1416 Facebook 1417 1 Hacker Way 1418 Menlo Park, CA 94025 1419 US 1421 Email: petr@fb.com 1423 Ariff Premji 1424 Arista Networks 1425 5453 Great America Parkway 1426 Santa Clara, CA 95054 1427 US 1429 Email: ariff@arista.com 1430 URI: http://arista.com/ 1432 Jon Mitchell (editor) 1433 Microsoft Corporation 1434 One Microsoft Way 1435 Redmond, WA 98052 1436 US 1438 Email: Jon.Mitchell@microsoft.com