idnits 2.17.00 (12 Aug 2021) /tmp/idnits65232/draft-ietf-teas-ietf-network-slices-09.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 document date (24 March 2022) is 58 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-07) exists of draft-ietf-opsawg-sap-02 == Outdated reference: A later version (-11) exists of draft-ietf-spring-nsh-sr-10 Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group A. Farrel, Ed. 3 Internet-Draft Old Dog Consulting 4 Intended status: Informational J. Drake, Ed. 5 Expires: 25 September 2022 Juniper Networks 6 R. Rokui 7 Ciena 8 S. Homma 9 NTT 10 K. Makhijani 11 Futurewei 12 L.M. Contreras 13 Telefonica 14 J. Tantsura 15 Microsoft 16 24 March 2022 18 Framework for IETF Network Slices 19 draft-ietf-teas-ietf-network-slices-09 21 Abstract 23 This document describes network slicing in the context of networks 24 built from IETF technologies. It defines the term "IETF Network 25 Slice" and establishes the general principles of network slicing in 26 the IETF context. 28 The document discusses the general framework for requesting and 29 operating IETF Network Slices, the characteristics of an IETF Network 30 Slice, the necessary system components and interfaces, and how 31 abstract requests can be mapped to more specific technologies. The 32 document also discusses related considerations with monitoring and 33 security. 35 This document also provides definitions of related terms to enable 36 consistent usage in other IETF documents that describe or use aspects 37 of IETF Network Slices. 39 Status of This Memo 41 This Internet-Draft is submitted in full conformance with the 42 provisions of BCP 78 and BCP 79. 44 Internet-Drafts are working documents of the Internet Engineering 45 Task Force (IETF). Note that other groups may also distribute 46 working documents as Internet-Drafts. The list of current Internet- 47 Drafts is at https://datatracker.ietf.org/drafts/current/. 49 Internet-Drafts are draft documents valid for a maximum of six months 50 and may be updated, replaced, or obsoleted by other documents at any 51 time. It is inappropriate to use Internet-Drafts as reference 52 material or to cite them other than as "work in progress." 54 This Internet-Draft will expire on 25 September 2022. 56 Copyright Notice 58 Copyright (c) 2022 IETF Trust and the persons identified as the 59 document authors. All rights reserved. 61 This document is subject to BCP 78 and the IETF Trust's Legal 62 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 63 license-info) in effect on the date of publication of this document. 64 Please review these documents carefully, as they describe your rights 65 and restrictions with respect to this document. Code Components 66 extracted from this document must include Revised BSD License text as 67 described in Section 4.e of the Trust Legal Provisions and are 68 provided without warranty as described in the Revised BSD License. 70 Table of Contents 72 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 73 1.1. Background . . . . . . . . . . . . . . . . . . . . . . . 4 74 2. Terms and Abbreviations . . . . . . . . . . . . . . . . . . . 5 75 2.1. Core Terminology . . . . . . . . . . . . . . . . . . . . 5 76 3. IETF Network Slice Objectives . . . . . . . . . . . . . . . . 7 77 3.1. Definition and Scope of IETF Network Slice . . . . . . . 7 78 3.2. IETF Network Slice Service . . . . . . . . . . . . . . . 8 79 3.2.1. Ancillary SDPs . . . . . . . . . . . . . . . . . . . 11 80 4. IETF Network Slice System Characteristics . . . . . . . . . . 12 81 4.1. Objectives for IETF Network Slices . . . . . . . . . . . 12 82 4.1.1. Service Level Objectives . . . . . . . . . . . . . . 13 83 4.1.2. Service Level Expectations . . . . . . . . . . . . . 14 84 4.2. IETF Network Slice Service Demarcation Points . . . . . . 17 85 4.3. IETF Network Slice Decomposition . . . . . . . . . . . . 19 86 5. Framework . . . . . . . . . . . . . . . . . . . . . . . . . . 20 87 5.1. IETF Network Slice Stakeholders . . . . . . . . . . . . . 20 88 5.2. Expressing Connectivity Intents . . . . . . . . . . . . . 20 89 5.3. IETF Network Slice Controller (NSC) . . . . . . . . . . . 22 90 5.3.1. IETF Network Slice Controller Interfaces . . . . . . 24 91 5.3.2. Management Architecture . . . . . . . . . . . . . . . 25 92 6. Realizing IETF Network Slices . . . . . . . . . . . . . . . . 26 93 6.1. Architecture to Realize IETF Network Slices . . . . . . . 27 94 6.2. Procedures to Realize IETF Network Slices . . . . . . . . 30 95 6.3. Applicability of ACTN to IETF Network Slices . . . . . . 31 96 6.4. Applicability of Enhanced VPNs to IETF Network Slices . . 31 97 6.5. Network Slicing and Aggregation in IP/MPLS Networks . . . 32 98 6.6. Network Slicing and Service Function Chaining (SFC) . . . 32 99 7. Isolation in IETF Network Slices . . . . . . . . . . . . . . 33 100 7.1. Isolation as a Service Requirement . . . . . . . . . . . 33 101 7.2. Isolation in IETF Network Slice Realization . . . . . . . 34 102 8. Management Considerations . . . . . . . . . . . . . . . . . . 34 103 9. Security Considerations . . . . . . . . . . . . . . . . . . . 34 104 10. Privacy Considerations . . . . . . . . . . . . . . . . . . . 35 105 11. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 36 106 12. Informative References . . . . . . . . . . . . . . . . . . . 36 107 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 40 108 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . 41 109 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 41 111 1. Introduction 113 A number of use cases benefit from network connections that, along 114 with connectivity, provide assurance of meeting a specific set of 115 objectives with respect to network resources use. This connectivity 116 and resource commitment is referred to as a network slice and is 117 expressed in terms of connectivity constructs (see Section 3) and 118 service objectives (see Section 4). Since the term network slice is 119 rather generic, the qualifying term "IETF" is used in this document 120 to limit the scope of network slice to network technologies described 121 and standardized by the IETF. This document defines the concept of 122 IETF Network Slices that provide connectivity coupled with a set of 123 specific commitments of network resources between a number of 124 endpoints (known as Service Demarcation Points (SDPs) - see 125 Section 2.1 and Section 4.2) over a shared underlay network. The 126 term IETF Network Slice service is also introduced to describe the 127 service requested by and provided to the service provider's customer. 129 Services that might benefit from IETF Network Slices include, but are 130 not limited to: 132 * 5G services (e.g. eMBB, URLLC, mMTC)(See [TS23501]) 134 * Network wholesale services 136 * Network infrastructure sharing among operators 138 * NFV connectivity and Data Center Interconnect 140 IETF Network Slices are created and managed within the scope of one 141 or more network technologies (e.g., IP, MPLS, optical). They are 142 intended to enable a diverse set of applications with different 143 requirements to coexist over a shared underlay network. A request 144 for an IETF Network Slice service is agnostic to the technology in 145 the underlay network so as to allow a customer to describe their 146 network connectivity objectives in a common format, independent of 147 the underlay technologies used. 149 This document also provides a framework for discussing IETF Network 150 Slices. The framework is intended as a structure for discussing 151 interfaces and technologies. It is not intended to specify a new set 152 of concrete interfaces or technologies. 154 For example, virtual private networks (VPNs) have served the industry 155 well as a means of providing different groups of users with logically 156 isolated access to a common network. The common or base network that 157 is used to support the VPNs is often referred to as an underlay 158 network, and the VPN is often called an overlay network. An overlay 159 network may, in turn, serve as an underlay network to support another 160 overlay network. 162 Note that it is conceivable that extensions to IETF technologies are 163 needed in order to fully support all the ideas that can be 164 implemented with network slices. Evaluation of existing 165 technologies, proposed extensions to existing protocols and 166 interfaces, and the creation of new protocols or interfaces is 167 outside the scope of this document. 169 1.1. Background 171 The concept of network slicing has gained traction driven largely by 172 needs surfacing from 5G ([NGMN-NS-Concept], [TS23501], and 173 [TS28530]). In [TS23501], a Network Slice is defined as "a logical 174 network that provides specific network capabilities and network 175 characteristics", and a Network Slice Instance is defined as "A set 176 of Network Function instances and the required resources (e.g. 177 compute, storage and networking resources) which form a deployed 178 Network Slice." According to [TS28530], an end-to-end network slice 179 consists of three major types of network segments: Radio Access 180 Network (RAN), Transport Network (TN) and Core Network (CN). An IETF 181 Network Slice provides the required connectivity between different 182 entities in RAN and CN segments of an end-to-end network slice, with 183 a specific performance commitment. For each end-to-end network 184 slice, the topology and performance requirement on a customer's use 185 of an IETF Network Slice can be very different, which requires the 186 underlay network to have the capability of supporting multiple 187 different IETF Network Slices. 189 While network slices are commonly discussed in the context of 5G, it 190 is important to note that IETF Network Slices are a narrower concept 191 with a broader usage profile, and focus primarily on particular 192 network connectivity aspects. Other systems, including 5G 193 deployments, may use IETF Network Slices as a component to create 194 entire systems and concatenated constructs that match their needs, 195 including end-to-end connectivity. 197 An IETF Network Slice could span multiple technologies and multiple 198 administrative domains. Depending on the IETF Network Slice 199 customer's requirements, an IETF Network Slice could be isolated from 200 other, often concurrent IETF Network Slices in terms of data, control 201 and management planes. 203 The customer expresses requirements for a particular IETF Network 204 Slice service by specifying what is required rather than how the 205 requirement is to be fulfilled. That is, the IETF Network Slice 206 customer's view of an IETF Network Slice is an abstract one. 208 Thus, there is a need to create logical network structures with 209 required characteristics. The customer of such a logical network can 210 require a degree of isolation and performance that previously might 211 not have been satisfied by overlay VPNs. Additionally, the IETF 212 Network Slice customer might ask for some level of control of their 213 virtual networks, e.g., to customize the service paths in a network 214 slice. 216 This document specifies definitions and a framework for the provision 217 of an IETF Network Slice service. Section 6 briefly indicates some 218 candidate technologies for realizing IETF Network Slices. 220 2. Terms and Abbreviations 222 The following abbreviations are used in this document. 224 * NSC: Network Slice Controller 226 * SLA: Service Level Agreement 228 * SLI: Service Level Indicator 230 * SLO: Service Level Objective 232 The meaning of these abbreviations is defined in greater details in 233 the remainder of this document. 235 2.1. Core Terminology 237 The following terms are presented here to give context. Other 238 terminology is defined in the remainder of this document. 240 Customer: A customer is the requester of an IETF Network Slice 241 service. Customers may request monitoring of SLOs. A customer 242 may be an entity such as an enterprise network or a network 243 operator, an individual working at such an entity, a private 244 individual contracting for a service, or an application or 245 software component. A customer may be an external party 246 (classically a paying customer) or a division of a network 247 operator that uses the service provided by another division of the 248 same operator. Other terms that have been applied to the customer 249 role are "client" and "consumer". 251 Provider: A provider is the organization that delivers an IETF 252 Network Slice service. A provider is the network operator that 253 controls the network resources used to construct the network slice 254 (that is, the network that is sliced). The provider's network 255 maybe a physical network or may be a virtual network supplied by 256 another service provider. 258 Customer Edge (CE): The customer device that provides connectivity 259 to a service provider. Examples include routers, Ethernet 260 switches, firewalls, 4G/5G RAN or Core nodes, application 261 accelerators, server load balancers, HTTP header enrichment 262 functions, and PEPs (Performance Enhancing Proxy). In some 263 circumstances CEs are provided to the customer and managed by the 264 provider. 266 Provider Edge (PE): The device within the provider network to which 267 a CE is attached. A CE may be attached to multiple PEs, and 268 multiple CEs may be attached to a given PE. 270 Attachment Circuit (AC): A channel connecting a CE and a PE over 271 which packets that belong to an IETF Network Slice service are 272 exchanged. An AC is, by definition, technology specific: that is, 273 the AC defines how customer traffic is presented to the provider 274 network. The customer and provider agree (through configuration) 275 on which values in which combination of layer 2 and layer 3 header 276 and payload fields within a packet identify to which {IETF Network 277 Slice service, connectivity construct, and SLOs/SLEs} that packet 278 is assigned. The customer and provider may agree on a per {IETF 279 Network Slice service, connectivity construct, and SLOs/SLEs} 280 basis to police or shape traffic on the AC in both the ingress (CE 281 to PE) direction and egress (PE to CE) direction, This ensures 282 that the traffic is within the capacity profile that is agreed in 283 an IETF Network Slice service. Excess traffic is dropped by 284 default, unless specific out-of-profile policies are agreed 285 between the customer and the provider. As described in 286 Section 4.2 the AC may be part of the IETF Network Slice service 287 or may be external to it. 289 Service Demarcation Point (SDP): The point at which an IETF Network 290 Slice service is delivered by a service provider to a customer. 291 Depending on the service delivery model (see Section 4.2) this may 292 be a CE or a PE, and could be a device, a software component, or 293 in the case of network functions virtualization (for example), be 294 an abstract function supported within the provider's network. 295 Each SDP must have a unique identifier (e.g., an IP address or MAC 296 address) within a given IETF Network Slice service and may use the 297 same identifier in multiple IETF Network Slice services. 299 An SDP may be abstracted as a Service Attachment Point (SAP) 300 [I-D.ietf-opsawg-sap] for the purpose generalizing the concept 301 across multiple service types and representing it in management 302 and configuration systems. 304 Connectivity Construct: A set of SDPs together with a communication 305 type that defines how traffic flows between the SDPs. An IETF 306 Network Slice service is specified in terms of a set of SDPs, the 307 associated connectivity constructs and the service objectives that 308 the customer wishes to see fulfilled. 310 3. IETF Network Slice Objectives 312 IETF Network Slices are created to meet specific requirements, 313 typically expressed as bandwidth, latency, latency variation, and 314 other desired or required characteristics. Creation of an IETF 315 Network Slice is initiated by a management system or other 316 application used to specify network-related conditions for particular 317 traffic flows in response to an actual or logical IETF Network Slice 318 service request. 320 Once created, these slices can be monitored, modified, deleted, and 321 otherwise managed. 323 Applications and components will be able to use these IETF Network 324 Slices to move packets between the specified end-points of the 325 service in accordance with specified characteristics. 327 3.1. Definition and Scope of IETF Network Slice 329 An IETF Network Slice service enables connectivity between a set of 330 Service Demarcation Points (SDPs) with specific Service Level 331 Objectives (SLOs) and Service Level Expectations (SLEs) over a common 332 underlay network. 334 An IETF Network Slice combines the connectivity resource requirements 335 and associated network capabilities such as bandwidth, latency, 336 jitter, and network functions with other resource behaviors such as 337 compute and storage availability. The definition of an IETF Network 338 Slice is independent of the connectivity and technologies used in the 339 underlay network. This allows an IETF Network Slice service customer 340 to describe their network connectivity and relevant objectives in a 341 common format, independent of the underlay technologies used. 343 IETF Network Slices may be combined hierarchically, so that a network 344 slice may itself be sliced. They may also be combined sequentially 345 so that various different networks can each be sliced and the network 346 slices placed into a sequence to provide an end-to-end service. This 347 form of sequential combination is utilized in some services such as 348 in 3GPP's 5G network [TS23501]. 350 An IETF Network Slice service is agnostic to the technology of the 351 underlay network, and its realization may be selected based upon 352 multiple considerations including its service requirements and the 353 capabilities of the underlay network. 355 The term "Slice" refers to a set of characteristics and behaviors 356 that differentiate one type of user-traffic from another. An IETF 357 Network Slice assumes that an underlay network is capable of changing 358 the configurations of the network devices on demand, through in-band 359 signaling or via controller(s) and fulfilling all or some of the 360 SLOs/SLEs to specific flows or to all of the traffic in the slice. 362 3.2. IETF Network Slice Service 364 A service provider delivers an IETF Network Slice service for a 365 customer. The IETF Network Slice service is specified in terms of a 366 set of SDPs, a set of one or more connectivity constructs between 367 subsets of these SDPs, and a set of SLOs and SLEs for each SDP 368 sending to each connectivity construct. A communication type (point- 369 to-point (P2P), point-to-multipoint (P2MP), or any-to-any (A2A)) is 370 specified for each connectivity construct. That is, in a given IETF 371 Network Slice service there may be one or more connectivity 372 constructs of the same or different type, each connectivity construct 373 may be between a different subset of SDPs, for a given connectivity 374 construct each sending SDP has its own set of SLOs and SLEs, and the 375 SLOs and SLEs in each set may be different. Note that a service 376 provider may decide how many connectivity constructs per IETF Network 377 Slice service it wishes to support such that an IETF Network Slice 378 service may be limited to one connectivity construct or may support 379 many. 381 This approach results in the following possible connectivity 382 constructs: 384 * For a P2P connectivity construct, there is one sending SDP and one 385 receiving SDP. This construct is like a private wire or a tunnel. 386 All traffic injected at the sending SDP is intended to be received 387 by the receiving SDP. The SLOs and SLEs apply at the sender (and 388 implicitly at the receiver). 390 * For a P2MP connectivity construct, there is only one sending SDP 391 and more than one receiving SDP. This is like a P2MP tunnel or 392 multi-access VLAN segment. All traffic from the sending SDP is 393 intended to be received by all the receiving SDPs. There is one 394 set of SLOs and SLEs that applies at the sending SDP (and 395 implicitly at all receiving SDPs). 397 * With an A2A connectivity construct, any sending SDP may send to 398 any one receiving SDP or any set of receiving SDPs in the 399 construct. There is an implicit level of routing in this 400 connectivity construct that is not present in the other 401 connectivity constructs because the provider's network must 402 determine to which receiving SDPs to deliver each packet. This 403 construct may be used to support P2P traffic between any pair of 404 SDPs, or to support multicast or broadcast traffic from one SDP to 405 a set of other SDPs. In the latter case, whether the service is 406 delivered using multicast within the provider's network or using 407 "ingress replication" or some other means is out of scope of the 408 specification of the service. A service provider may choose to 409 support A2A constructs, but to limit the traffic to unicast. 411 The SLOs/SLEs in an A2A connectivity construct apply to individual 412 sending SDPs regardless of the receiving SDPs, and there is no 413 linkage between sender and receiver in the specification of the 414 connectivity construct. A sending SDP may be "disappointed" if 415 the receiver is over-subscribed. If a customer wants to be more 416 specific about different behaviors from one SDP to another SDP, 417 they should use P2P connectivity constructs. 419 A customer traffic flow may be unicast or multicast, and various 420 network realizations are possible: 422 * Unicast traffic may be mapped to a P2P connectivity construct for 423 direct delivery, or to an A2A connectivity construct for the 424 service provider to perform routing to the destination SDP. It 425 would be unusual to use a P2MP connectivity construct to deliver 426 unicast traffic because all receiving SDPs would get a copy, but 427 this can still be done if the receivers are capable of dropping 428 the unwanted traffic. 430 * A bidirectional unicast service can be constructed by specifying 431 two P2P connectivity constructs. An additional SLE may specify 432 fate-sharing in this case. 434 * Multicast traffic may be mapped to a set of P2P connectivity 435 constructs, a single P2MP connectivity construct, or a mixture of 436 P2P and P2MP connectivity constructs. Multicast may also be 437 supported by an A2A connectivity construct. The choice clearly 438 influences how and where traffic is replicated in the network. 439 With a P2MP or A2A connectivity construct, it is the operator's 440 choice whether to realize the construct with ingress replication, 441 multicast in the core, P2MP tunnels, or hub-and-spoke. This 442 choice should not change how the customer perceives the service. 444 * The concept of a multipoint-to-point (MP2P) service can be 445 realized with multiple P2P connectivity constructs. Note that, in 446 this case, the egress may simultaneously receive traffic from all 447 ingresses. The SLOs at the sending SDPs must be set with this in 448 mind because the provider's network is not capable of coordinating 449 the policing of traffic across multiple distinct source SDPs. It 450 is assumed that the customer, requesting SLOs for the various P2P 451 connectivity constructs, is aware of the capabilities of the 452 receiving SDP. If the receiver receives more traffic than it can 453 handle, it may drop some and introduce queuing delays. 455 * The concept of a multipoint-to-multipoint (MP2MP) service can best 456 be realized using a set of P2MP connectivity constructs, but could 457 be delivered over an A2A connectivity construct if each sender is 458 using multicast. As with MP2P, the customer is assumed to be 459 familiar with the capabilities of all receivers. A customer may 460 wish to achieve an MP2MP service using a hub-and-spoke 461 architecture where they control the hub: that is, the hub may be 462 an SDP or an ancillary SDP (see Section 3.2.1) and the service may 463 be achieved by using a set of P2P connectivity constructs to the 464 hub, and a single P2MP connectivity construct from the hub. 466 From the above, it can be seen that the SLOs of the senders define 467 the SLOs for the receivers on any connectivity construct. That is, 468 and in particular, the network may be expected to handle the traffic 469 volume from a sender to all destinations. This extends to all 470 connectivity constructs in an IETF Network Slice service. 472 Note that the realization of an IETF Network Slice service does not 473 need to map the connectivity constructs one-to-one onto underlying 474 network constructs (such as tunnels, etc.). The service provided to 475 the customer is distinct from how the provider decides to deliver 476 that service. 478 If a CE has multiple attachment circuits to a PE within a given IETF 479 Network Slice service and they are operating in single-active mode, 480 then all traffic between the CE and its attached PEs transits a 481 single attachment circuit; if they are operating in in all-active 482 mode, then traffic between the CE and its attached PEs is distributed 483 across all of the active attachment circuits. 485 A given sending SDP may be part of multiple connectivity constructs 486 within a single IETF Network Slice service, and the SDP may have 487 different SLOs and SLEs for each connectivity construct to which it 488 is sending. Note that a given sending SDP's SLOs and SLEs for a 489 given connectivity construct apply between it and each of the 490 receiving SDPs for that connectivity construct. 492 An IETF Network Slice service provider may freely make a deployment 493 choice as to whether to offer a 1:1 relationship between IETF Network 494 Slice service and connectivity construct, or to support multiple 495 connectivity constructs in a single IETF Network Slice service. In 496 the former case, the provider might need to deliver multiple IETF 497 Network Slice services to achieve the function of the second case. 499 It should be noted that per Section 9 of [RFC4364] an IETF Network 500 Slice service customer may actually provide IETF Network Slice 501 services to other customers in a mode sometimes referred to as 502 "carrier's carrier". In this case, the underlying IETF Network Slice 503 service provider may be owned and operated by the same or a different 504 provider network. As noted in Section 4.3, network slices may be 505 composed hierarchically or serially. 507 Section 4.2 provides a description of endpoints in the context of 508 IETF network slicing. These are known as Service Demarcation Points 509 (SDPs). For a given IETF Network Slice service, the customer and 510 provider agree, on a per-SDP basis which end of the attachment 511 circuit provides the SDP (i.e., whether the attachment circuit is 512 inside or outside the IETF Network Slice service). This determines 513 whether the attachment circuit is subject to the set of SLOs and SLEs 514 at the specific SDP. 516 3.2.1. Ancillary SDPs 518 It may be the case that the set of SDPs needs to be supplemented with 519 additional senders or receivers. An additional sender could be, for 520 example, an IPTV or DNS server either within the provider's network 521 or attached to it, while an extra receiver could be, for example, a 522 node reachable via the Internet. This is modelled as a set of 523 ancillary SDPs which supplement the other SDPs in one or more 524 connectivity constructs, or which have their own connectivity 525 constructs. Note that an ancillary SDP can either have a resolvable 526 address, e.g., an IP address or MAC address, or the SDP may be a 527 placeholder, e.g., IPTV or DNS server, which is resolved within the 528 provider's network when the IETF Network Slice service is 529 instantiated. 531 4. IETF Network Slice System Characteristics 533 The following subsections describe the characteristics of IETF 534 Network Slices in addition to the list of SDPs, the connectivity 535 constructs, and the technology of the ACs. 537 4.1. Objectives for IETF Network Slices 539 An IETF Network Slice service is defined in terms of quantifiable 540 characteristics known as Service Level Objectives (SLOs) and 541 unquantifiable characteristics known as Service Level Expectations 542 (SLEs). SLOs are expressed in terms Service Level Indicators (SLIs), 543 and together with the SLEs form the contractual agreement between 544 service customer and service provider known as a Service Level 545 Agreement (SLA). 547 The terms are defined as follows: 549 * A Service Level Indicator (SLI) is a quantifiable measure of an 550 aspect of the performance of a network. For example, it may be a 551 measure of throughput in bits per second, or it may be a measure 552 of latency in milliseconds. 554 * A Service Level Objective (SLO) is a target value or range for the 555 measurements returned by observation of an SLI. For example, an 556 SLO may be expressed as "SLI <= target", or "lower bound <= SLI <= 557 upper bound". A customer can determine whether the provider is 558 meeting the SLOs by performing measurements on the traffic. 560 * A Service Level Expectation (SLE) is an expression of an 561 unmeasurable service-related request that a customer of an IETF 562 Network Slice makes of the provider. An SLE is distinct from an 563 SLO because the customer may have little or no way of determining 564 whether the SLE is being met, but they still contract with the 565 provider for a service that meets the expectation. 567 * A Service Level Agreement (SLA) is an explicit or implicit 568 contract between the customer of an IETF Network Slice service and 569 the provider of the slice. The SLA is expressed in terms of a set 570 of SLOs and SLEs that are to be applied for a given connectivity 571 construct between a sending SDP and the set of receiving SDPs, and 572 may describe the extent to which divergence from individual SLOs 573 and SLEs can be tolerated, and commercial terms as well as any 574 consequences for violating these SLOs and SLEs. 576 4.1.1. Service Level Objectives 578 SLOs define a set of measurable network attributes and 579 characteristics that describe an IETF Network Slice service. SLOs do 580 not describe how an IETF Network Slice service is implemented or 581 realized in the underlying network layers. Instead, they are defined 582 in terms of dimensions of operation (time, capacity, etc.), 583 availability, and other attributes. 585 An IETF Network Slice service may include multiple connectivity 586 constructs that associate sets of endpoints (SDPs). SLOs apply to a 587 given connectivity construct and apply to a specific direction of 588 traffic flow. That is, they apply to a specific sending SDP and the 589 connection to the specific set of receiving SDPs. 591 The SLOs are combined with Service Level Expectations in an SLA. 593 4.1.1.1. Some Common SLOs 595 SLOs can be described as 'Directly Measurable Objectives': they are 596 always measurable. See Section 4.1.2 for the description of Service 597 Level Expectations which are unmeasurable service-related requests 598 sometimes known as 'Indirectly Measurable Objectives'. 600 Objectives such as guaranteed minimum bandwidth, guaranteed maximum 601 latency, maximum permissible delay variation, maximum permissible 602 packet loss rate, and availability are 'Directly Measurable 603 Objectives'. Future specifications (such as IETF Network Slice 604 service YANG models) may precisely define these SLOs, and other SLOs 605 may be introduced as described in Section 4.1.1.2. 607 The definition of these objectives are as follows: 609 Guaranteed Minimum Bandwidth: Minimum guaranteed bandwidth between 610 two endpoints at any time. The bandwidth is measured in data rate 611 units of bits per second and is measured unidirectionally. 613 Guaranteed Maximum Latency: Upper bound of network latency when 614 transmitting between two endpoints. The latency is measured in 615 terms of network characteristics (excluding application-level 616 latency). [RFC7679] discusses one-way metrics. 618 Maximum Permissible Delay Variation: Packet delay variation (PDV) as 619 defined by [RFC3393], is the difference in the one-way delay 620 between sequential packets in a flow. This SLO sets a maximum 621 value PDV for packets between two endpoints. 623 Maximum Permissible Packet Loss Rate: The ratio of packets dropped 624 to packets transmitted between two endpoints over a period of 625 time. See [RFC7680]. 627 Availability: The ratio of uptime to the sum of uptime and downtime, 628 where uptime is the time the connectivity construct is available 629 in accordance with all of the SLOs associated with it. 631 4.1.1.2. Other Service Level Objectives 633 Additional SLOs may be defined to provide additional description of 634 the IETF Network Slice service that a customer requests. These would 635 be specified in further documents. 637 If the IETF Network Slice service is traffic aware, other traffic 638 specific characteristics may be valuable including MTU, traffic-type 639 (e.g., IPv4, IPv6, Ethernet or unstructured), or a higher-level 640 behavior to process traffic according to user-application (which may 641 be realized using network functions). 643 4.1.2. Service Level Expectations 645 SLEs define a set of network attributes and characteristics that 646 describe an IETF Network Slice service, but which are not directly 647 measurable by the customer. Even though the delivery of an SLE 648 cannot usually be determined by the customer, the SLEs form an 649 important part of the contract between customer and provider. 651 Quite often, an SLE will imply some details of how an IETF Network 652 Slice service is realized by the provider, although most aspects of 653 the implementation in the underlying network layers remain a free 654 choice for the provider. For example, activating unicast or 655 multicast capabilities to deliver an IETF Network Slice service could 656 be explicitly requested by a customer or could be left as an 657 engineering decision for the service provider based on capabilities 658 of the network and operational choices. 660 SLEs may be seen as aspirational on the part of the customer, and 661 they are expressed as behaviors that the provider is expected to 662 apply to the network resources used to deliver the IETF Network Slice 663 service. Of course, over time, it is possible that mechanisms will 664 be developed that enable a customer to verify the provision of an 665 SLE, at which point it effectively becomes an SLO. The SLEs are 666 combined with SLOs in an SLA. 668 An IETF Network Slice service may include multiple connectivity 669 constructs that associate sets of endpoints (SDPs). SLEs apply to a 670 given connectivity construct and apply to specific directions of 671 traffic flow. That is, they apply to a specific sending SDP and the 672 connection to the specific set of receiving SDPs. However, being 673 more general in nature than SLOs, SLEs may commonly be applied to all 674 connectivity constructs in an IETF Network Slice service. 676 4.1.2.1. Some Common SLEs 678 SLEs can be described as 'Indirectly Measurable Objectives': they are 679 not generally directly measurable by the customer. 681 Security, geographic restrictions, maximum occupancy level, and 682 isolation are example SLEs as follows. 684 Security: A customer may request that the provider applies 685 encryption or other security techniques to traffic flowing between 686 SDPs of a connectivity construct within an IETF Network Slice 687 service. For example, the customer could request that only 688 network links that have MACsec [MACsec] enabled are used to 689 realize the connectivity construct. 691 This SLE may include a request for encryption (e.g., [RFC4303]) 692 between the two SDPs explicitly to meet the architectural 693 recommendations in [TS33.210] or for compliance with [HIPAA] or 694 [PCI]. 696 Whether or not the provider has met this SLE is generally not 697 directly observable by the customer and cannot be measured as a 698 quantifiable metric. 700 Please see further discussion on security in Section 9. 702 Geographic Restrictions: A customer may request that certain 703 geographic limits are applied to how the provider routes traffic 704 for the IETF Network Slice service. For example, the customer may 705 have a preference that its traffic does not pass through a 706 particular country for political or security reasons. 708 Whether or not the provider has met this SLE is generally not 709 directly observable by the customer and cannot be measured as a 710 quantifiable metric. 712 Maximal Occupancy Level: The maximal occupancy level specifies the 713 number of flows to be admitted and optionally a maximum number of 714 countable resource units (e.g., IP or MAC addresses) an IETF 715 Network Slice service can consume. Since an IETF Network Slice 716 service may include multiple connectivity constructs, this SLE 717 should also say whether it applies for the entire IETF Network 718 Slice service, for group of connections, or on a per connection 719 basis. 721 Again, a customer may not be able to fully determine whether this 722 SLE is being met by the provider. 724 Isolation: As described in Section 7, a customer may request that 725 its traffic within its IETF Network Slice service is isolated from 726 the effects of other network services supported by the same 727 provider. That is, if another service exceeds capacity or has a 728 burst of traffic, the customer's IETF Network Slice service should 729 remain unaffected and there should be no noticeable change to the 730 quality of traffic delivered. 732 In general, a customer cannot tell whether a service provider is 733 meeting this SLE. They cannot tell whether the variation of an 734 SLI is because of changes in the underlay network or because of 735 interference from other services carried by the network. If the 736 service varies within the allowed bounds of the SLOs, there may be 737 no noticeable indication that this SLE has been violated. 739 Diversity: A customer may request that different connectivity 740 constructs use different underlay network resources. This might 741 be done to enhance the availability of the connectivity constructs 742 within an IETF Network Slice service. 744 While availability is a measurable objective (see Section 4.1.1.1) 745 this SLE requests a finer grade of control and is not directly 746 measurable (although the customer might become suspicious if two 747 connectivity constructs fail at the same time). 749 4.2. IETF Network Slice Service Demarcation Points 751 As noted in Section 3.1, an IETF Network Slice provides connectivity 752 between sets of SDPs with specific SLOs and SLEs. Section 3.2 goes 753 on to describe how the IETF Network Slice service is composed of a 754 set of one or more connectivity constructs that describe connectivity 755 between the Service Demarcation Points (SDPs) across the underlay 756 network. 758 The characteristics of IETF Network Slice SDPs are as follows. 760 * SDPs are conceptual points of connection to an IETF Network Slice. 761 As such, they serve as the IETF Network Slice ingress/egress 762 points. 764 * Each SDP maps to a device, application, or a network function, 765 such as (but not limited to) routers, switches, interfaces/ports, 766 firewalls, WAN, 4G/5G RAN nodes, 4G/5G Core nodes, application 767 accelerators, server load balancers, NAT44 [RFC3022], NAT64 768 [RFC6146], HTTP header enrichment functions, and Performance 769 Enhancing Proxies (PEPs) [RFC3135]. 771 * An SDP is identified by a unique identifier in the context of an 772 IETF Network Slice customer. 774 * The provider associates each SDP with a set of provider-scope 775 identifiers such as IP addresses, encapsulation-specific 776 identifiers (e.g., VLAN tag, MPLS Label), interface/port numbers, 777 node ID, etc. 779 * SDPs are mapped to endpoints of services/tunnels/paths within the 780 IETF Network Slice during its initialization and realization. 782 - A combination of the SDP identifier and SDP provider-network- 783 scope identifiers define an SDP in the context of the Network 784 Slice Controller (NSC) (see Section 5.3). 786 - The NSC will use the SDP provider-network-scope identifiers as 787 part of the process of realizing the IETF Network Slice. 789 For a given IETF Network Slice service, the IETF Network Slice 790 customer and provider agree where the endpoint (i.e., the service 791 demarcation point) is located. This determines what resources at the 792 edge of the network form part of the IETF Network Slice and are 793 subject to the set of SLOs and SLEs for a specific endpoint. 795 Figure 1 shows different potential scopes of an IETF Network Slice 796 that are consistent with the different SDP locations. For the 797 purpose of this discussion and without loss of generality, the figure 798 shows customer edge (CE) and provider edge (PE) nodes connected by 799 attachment circuits (ACs). Notes after the figure give some 800 explanations. 802 |<---------------------- (1) ---------------------->| 803 | | 804 | |<-------------------- (2) -------------------->| | 805 | | | | 806 | | |<----------- (3) ----------->| | | 807 | | | | | | 808 | | | |<-------- (4) -------->| | | | 809 | | | | | | | | 810 V V AC V V V V AC V V 811 +-----+ | +-----+ +-----+ | +-----+ 812 | |--------| | | |--------| | 813 | CE1 | | | PE1 |. . . . . . . . .| PE2 | | | CE2 | 814 | |--------| | | |--------| | 815 +-----+ | +-----+ +-----+ | +-----+ 816 ^ ^ ^ ^ 817 | | | | 818 | | | | 819 Customer Provider Provider Customer 820 Edge 1 Edge 1 Edge 2 Edge 2 822 Figure 1: Positioning IETF Service Demarcation Points 824 Explanatory notes for Figure 1 are as follows: 826 1. If the CE is operated by the IETF Network Slice service provider, 827 then the edge of the IETF Network Slice may be within the CE. In 828 this case the slicing process may utilize resources from within 829 the CE such as buffers and queues on the outgoing interfaces. 831 2. The IETF Network Slice may be extended as far as the CE, to 832 include the AC, but not to include any part of the CE. In this 833 case, the CE may be operated by the customer or the provider. 834 Slicing the resources on the AC may require the use of traffic 835 tagging (such as through Ethernet VLAN tags) or may require 836 traffic policing at the AC link ends. 838 3. In another model, the SDPs of the IETF Network Slice are the 839 customer-facing ports on the PEs. This case can be managed in a 840 way that is similar to a port-based VPN: each port (AC) or 841 virtual port (e.g., VLAN tag) identifies the IETF Network Slice 842 and maps to an IETF Network Slice SDP. 844 4. Finally, the SDP may be within the PE. In this mode, the PE 845 classifies the traffic coming from the AC according to 846 information (such as the source and destination IP addresses, 847 payload protocol and port numbers, etc.) in order to place it 848 onto an IETF Network Slice. 850 The choice of which of these options to apply is entirely up to the 851 network operator. It may limit or enable the provisioning of 852 particular managed services and the operator will want to consider 853 how they want to manage CEs and what control they wish to offer the 854 customer over AC resources. 856 Note that Figure 1 shows a symmetrical positioning of SDPs, but this 857 decision can be taken on a per-SDP basis through agreement between 858 the customer and provider. 860 In practice, it may be necessary to map traffic not only onto an IETF 861 Network Slice, but also onto a specific connectivity construct if the 862 IETF Network Slice supports more than one with a source at the 863 specific SDP. The mechanism used will be one of the mechanisms 864 described above, dependent on how the SDP is realized. 866 Finally, note (as described in Section 2.1) that an SDP is an 867 abstract endpoint of an IETF Network Slice service and as such may be 868 a device, interface, or software component and may, in the case of 869 network functions virtualization (for example), be an abstract 870 function supported within the provider's network. 872 4.3. IETF Network Slice Decomposition 874 Operationally, an IETF Network Slice may be composed of two or more 875 IETF Network Slices as specified below. Decomposed network slices 876 are independently realized and managed. 878 * Hierarchical (i.e., recursive) composition: An IETF Network Slice 879 can be further sliced into other network slices. Recursive 880 composition allows an IETF Network Slice at one layer to be used 881 by the other layers. This type of multi-layer vertical IETF 882 Network Slice associates resources at different layers. 884 * Sequential composition: Different IETF Network Slices can be 885 placed into a sequence to provide an end-to-end service. In 886 sequential composition, each IETF Network Slice would potentially 887 support different dataplanes that need to be stitched together. 889 5. Framework 891 A number of IETF Network Slice services will typically be provided 892 over a shared underlay network infrastructure. Each IETF Network 893 Slice consists of both the overlay connectivity and a specific set of 894 dedicated network resources and/or functions allocated in a shared 895 underlay network to satisfy the needs of the IETF Network Slice 896 customer. In at least some examples of underlay network 897 technologies, the integration between the overlay and various 898 underlay resources is needed to ensure the guaranteed performance 899 requested for different IETF Network Slices. 901 5.1. IETF Network Slice Stakeholders 903 An IETF Network Slice and its realization involves the following 904 stakeholders. The IETF Network Slice customer and IETF Network Slice 905 provider (see Section 2.1) are also stakeholders. 907 Orchestrator: An orchestrator is an entity that composes different 908 services, resource, and network requirements. It interfaces with 909 the IETF NSC when composing a complex service such as an end-to- 910 end network slice. 912 IETF Network Slice Controller (NSC): The NSC realizes an IETF 913 Network Slice in the underlay network, and maintains and monitors 914 the run-time state of resources and topologies associated with it. 915 A well-defined interface is needed to support interworking between 916 different NSC implementations and different orchestrator 917 implementations. 919 Network Controller: The Network Controller is a form of network 920 infrastructure controller that offers network resources to the NSC 921 to realize a particular network slice. This may be an existing 922 network controller associated with one or more specific 923 technologies that may be adapted to the function of realizing IETF 924 Network Slices in a network. 926 5.2. Expressing Connectivity Intents 928 An IETF Network Slice customer communicates with the NSC using the 929 IETF Network Slice Service Interface. 931 An IETF Network Slice customer may be a network operator who, in 932 turn, use the IETF Network Slice to provide a service for another 933 IETF Network Slice customer. 935 Using the IETF Network Slice Service Interface, a customer expresses 936 requirements for a particular slice by specifying what is required 937 rather than how that is to be achieved. That is, the customer's view 938 of a slice is an abstract one. Customers normally have limited (or 939 no) visibility into the provider network's actual topology and 940 resource availability information. 942 This should be true even if both the customer and provider are 943 associated with a single administrative domain, in order to reduce 944 the potential for adverse interactions between IETF Network Slice 945 customers and other users of the underlay network infrastructure. 947 The benefits of this model can include the following. 949 * Security: The underlay network components are less exposed to 950 attack because the underlay network (or network operator) does not 951 need to expose network details (topology, capacity, etc.) to the 952 IETF Network Slice customers. 954 * Layered Implementation: The underlay network comprises network 955 elements that belong to a different layer network than customer 956 applications. Network information (advertisements, protocols, 957 etc.) that a customer cannot interpret or respond to is not 958 exposed to the customer. (Note - a customer should not use 959 network information not exposed via the IETF Network Slice Service 960 Interface, even if that information is available.) 962 * Scalability: Customers do not need to know any information beyond 963 that which is exposed via the IETF Network Slice Service 964 Interface. 966 The general issues of abstraction in a TE network are described more 967 fully in [RFC7926]. 969 This framework document does not assume any particular technology 970 layer at which IETF Network Slices operate. A number of layers 971 (including virtual L2, Ethernet or, IP connectivity) could be 972 employed. 974 Data models and interfaces are needed to set up IETF Network Slices, 975 and specific interfaces may have capabilities that allow creation of 976 slices within specific technology layers. 978 Layered virtual connections are comprehensively discussed in other 979 IETF documents. See, for instance, GMPLS-based networks [RFC5212] 980 and [RFC4397], or Abstraction and Control of TE Networks (ACTN) 981 [RFC8453] and [RFC8454]. The principles and mechanisms associated 982 with layered networking are applicable to IETF Network Slices. 984 There are several IETF-defined mechanisms for expressing the need for 985 a desired logical network. The IETF Network Slice Service Interface 986 carries data either in a protocol-defined format, or in a formalism 987 associated with a modeling language. 989 For instance: 991 * The Path Computation Element (PCE) Communication Protocol (PCEP) 992 [RFC5440] and GMPLS User-Network Interface (UNI) using RSVP-TE 993 [RFC4208] use a TLV-based binary encoding to transmit data. 995 * The Network Configuration Protocol (NETCONF) [RFC6241] and 996 RESTCONF Protocol [RFC8040] use XML and JSON encoding. 998 * gRPC/GNMI [I-D.openconfig-rtgwg-gnmi-spec] uses a binary encoded 999 programmable interface. ProtoBufs can be used to model gRPC and 1000 GNMI data. 1002 * For data modeling, YANG ([RFC6020] and [RFC7950]) may be used to 1003 model configuration and other data for NETCONF, RESTCONF, and 1004 GNMI, among others. 1006 While several generic formats and data models for specific purposes 1007 exist, it is expected that IETF Network Slice management may require 1008 enhancement or augmentation of existing data models. Further, it is 1009 possible that mechanisms will be needed to determine the feasibility 1010 of service requests before they are actually made. 1012 5.3. IETF Network Slice Controller (NSC) 1014 The IETF NSC takes abstract requests for IETF Network Slices and 1015 implements them using a suitable underlay technology. An IETF NSC is 1016 the key component for control and management of the IETF Network 1017 Slice. It provides the creation/modification/deletion, monitoring 1018 and optimization of IETF Network Slices in a multi-domain, a multi- 1019 technology and multi-vendor environment. 1021 The main task of the IETF NSC is to map abstract IETF Network Slice 1022 requirements to concrete technologies and establish required 1023 connectivity ensuring that resources are allocated to the IETF 1024 Network Slice as necessary. 1026 The IETF Network Slice Service Interface is used for communicating 1027 details of an IETF Network Slice (configuration, selected policies, 1028 operational state, etc.), as well as information about status and 1029 performance of the IETF Network Slice. The details for this IETF 1030 Network Slice Service Interface are not in scope for this document. 1032 The controller provides the following functions. 1034 * Provides an IETF Network Slice Service Interface for 1035 creation/modification/deletion of the IETF Network Slices that is 1036 agnostic to the technology of the underlay network. The API 1037 exposed by this interface communicates the Service Demarcation 1038 Points of the IETF Network Slice, IETF Network Slice SLO/SLE 1039 parameters (and possibly monitoring thresholds), applicable input 1040 selection (filtering) and various policies, and provides a way to 1041 monitor the slice. 1043 * Determines an abstract topology connecting the SDPs of the IETF 1044 Network Slice that meets criteria specified via the IETF Network 1045 Slice Service Interface. The NSC also retains information about 1046 the mapping of this abstract topology to underlay components of 1047 the IETF Network Slice as necessary to monitor IETF Network Slice 1048 status and performance. 1050 * Provides "Mapping Functions" for the realization of IETF Network 1051 Slices. In other words, it will use the mapping functions that: 1053 - map IETF Network Slice Service Interface requests that are 1054 agnostic to the technology of the underlay network to 1055 technology-specific network configuration interfaces. 1057 - map filtering/selection information as necessary to entities in 1058 the underlay network. 1060 * The controller collects telemetry data (e.g., OAM results, 1061 statistics, states, etc.) via a network configuration interface 1062 for all elements in the abstract topology used to realize the IETF 1063 Network Slice. 1065 * Evaluates the current performance against IETF Network Slice SLO 1066 parameters using the telemetry data from the underlying 1067 realization of an IETF Network Slice (i.e., services/paths/ 1068 tunnels). Exposes this performance to the IETF Network Slice 1069 customer via the IETF Network Slice Service Interface. The IETF 1070 Network Slice Service Interface may also include the capability to 1071 provide notifications if the IETF Network Slice performance 1072 reaches threshold values defined by the IETF Network Slice 1073 customer. 1075 5.3.1. IETF Network Slice Controller Interfaces 1077 The interworking and interoperability among the different 1078 stakeholders to provide common means of provisioning, operating and 1079 monitoring the IETF Network Slices is enabled by the following 1080 communication interfaces (see Figure 2). 1082 IETF Network Slice Service Interface: The IETF Network Slice Service 1083 Interface is an interface between a customer's higher level 1084 operation system (e.g., a network slice orchestrator or a customer 1085 network management system) and the NSC. It is agnostic to the 1086 technology of the underlay network. The customer can use this 1087 interface to communicate the requested characteristics and other 1088 requirements for the IETF Network Slice, and the NSC can use the 1089 interface to report the operational state of an IETF Network Slice 1090 to the customer. 1092 Network Configuration Interface: The Network Configuration Interface 1093 is an interface between the NSC and network controllers. It is 1094 technology-specific and may be built around the many network 1095 models already defined within the IETF. 1097 These interfaces can be considered in the context of the Service 1098 Model and Network Model described in [RFC8309] and, together with the 1099 Device Configuration Interface used by the Network Controllers, 1100 provides a consistent view of service delivery and realization. 1102 +------------------------------------------+ 1103 | Customer higher level operation system | 1104 | (e.g E2E network slice orchestrator, | 1105 | customer network management system) | 1106 +------------------------------------------+ 1107 A 1108 | IETF Network Slice Service Interface 1109 V 1110 +------------------------------------------+ 1111 | IETF Network Slice Controller (NSC) | 1112 +------------------------------------------+ 1113 A 1114 | Network Configuration Interface 1115 V 1116 +------------------------------------------+ 1117 | Network Controllers | 1118 +------------------------------------------+ 1120 Figure 2: Interfaces of the IETF Network Slice Controller 1122 5.3.1.1. IETF Network Slice Service Interface 1124 The IETF Network Slice Controller provides an IETF Network Slice 1125 Service Interface that allows customers to request and monitor IETF 1126 Network Slices. Customers operate on abstract IETF Network Slices, 1127 with details related to their realization hidden. 1129 The IETF Network Slice Service Interface is also independent of the 1130 type of network functions or services that need to be connected, 1131 i.e., it is independent of any specific storage, software, protocol, 1132 or platform used to realize physical or virtual network connectivity 1133 or functions in support of IETF Network Slices. 1135 The IETF Network Slice Service Interface uses protocol mechanisms and 1136 information passed over those mechanisms to convey desired attributes 1137 for IETF Network Slices and their status. The information is 1138 expected to be represented as a well-defined data model, and should 1139 include at least SDP and connectivity information, SLO/SLE 1140 specification, and status information. 1142 5.3.2. Management Architecture 1144 The management architecture described in Figure 2 may be further 1145 decomposed as shown in Figure 3. This should also be seen in the 1146 context of the component architecture shown in Figure 4 and 1147 corresponds to the architecture in [RFC8309]. 1149 -------------- 1150 | Network | 1151 | Slice | 1152 | Orchestrator | 1153 -------------- 1154 | IETF Network Slice 1155 | Service Request 1156 | Customer view 1157 ....|................................ 1158 -v------------------- Operator view 1159 |Controller | 1160 | ------------ | 1161 | | IETF | | 1162 | | Network | |--> Virtual Network 1163 | | Slice | | 1164 | | Controller | | 1165 | | (NSC) | | 1166 | ------------ | 1167 ..| | Network |............ 1168 | | Configuration | Underlay Network 1169 | v | 1170 | ------------ | 1171 | | Network | | 1172 | | Controller | | 1173 | | (NC) | | 1174 | ------------ | 1175 --------------------- 1176 | Device Configuration 1177 v 1179 Figure 3: Interface of IETF Network Slice Management Architecture 1181 6. Realizing IETF Network Slices 1183 Realization of IETF Network Slices is out of scope of this document. 1184 It is a mapping of the definition of the IETF Network Slice to the 1185 underlying infrastructure and is necessarily technology-specific and 1186 achieved by the NSC over the Network Configuration Interface. 1187 However, this section provides an overview of the components and 1188 processes involved in realizing an IETF Network Slice. 1190 The realization can be achieved in a form of either physical or 1191 logical connectivity using VPNs, virtual networks (VNs), or a variety 1192 of tunneling technologies such as Segment Routing, MPLS, etc. 1193 Accordingly, SDPs may be realized as physical or logical service or 1194 network functions. 1196 6.1. Architecture to Realize IETF Network Slices 1198 The architecture described in this section is deliberately at a high 1199 level. It is not intended to be prescriptive: implementations and 1200 technical solutions may vary freely. However, this approach provides 1201 a common framework that other documents may reference in order to 1202 facilitate a shared understanding of the work. 1204 Figure 4 shows the architectural components of a network managed to 1205 provide IETF Network Slices. The customer's view is of individual 1206 IETF Network Slices with their SDPs, and connectivity constructs. 1207 Requests for IETF Network Slices are delivered to the NSC. 1209 The figure shows, without loss of generality, the CEs, ACs, and PEs, 1210 that exist in the network. The SDPs are not shown and can be placed 1211 in any of the ways described in Section 4.2. 1213 -- -- -- 1214 |CE| |CE| |CE| 1215 -- -- -- 1216 AC : AC : AC : 1217 ---------------------- ------- 1218 ( |PE|....|PE|....|PE| ) ( IETF ) 1219 IETF Network ( --: -- :-- ) ( Network ) 1220 Slice Service ( :............: ) ( Slice ) 1221 Request ( IETF Network Slice ) ( ) Customer 1222 v ---------------------- ------- View 1223 v ............................\........./............... 1224 v \ / Provider 1225 v >>>>>>>>>>>>>>> Grouping/Mapping v v View 1226 v ^ ----------------------------------------- 1227 v ^ ( |PE|.......|PE|........|PE|.......|PE| ) 1228 --------- ( --: -- :-- -- ) 1229 | | ( :...................: ) 1230 | NSC | ( Network Resource Partition ) 1231 | | ----------------------------------------- 1232 | | ^ 1233 | |>>>>> Resource Partitioning | 1234 --------- of Filter Topology | 1235 v v | 1236 v v ----------------------------- -------- 1237 v v (|PE|..-..|PE|... ..|PE|..|PE|) ( ) 1238 v v ( :-- |P| -- :-: -- :-- ) ( Filter ) 1239 v v ( :.- -:.......|P| :- ) ( Topology ) 1240 v v ( |P|...........:-:.......|P| ) ( ) 1241 v v ( - Filter Topology ) -------- 1242 v v ----------------------------- ^ 1243 v >>>>>>>>>>>> Topology Filter ^ / 1244 v ...........................\............../........... 1245 v \ / Underlay 1246 ---------- \ / (Physical) 1247 | | \ / Network 1248 | Network | ---------------------------------------------- 1249 |Controller| ( |PE|.....-.....|PE|...... |PE|.......|PE| ) 1250 | | ( -- |P| -- :-...:-- -..:-- ) 1251 ---------- ( : -:.............|P|.........|P| ) 1252 v ( -......................:-:..- - ) 1253 >>>>>>> ( |P|.........................|P|......: ) 1254 Program the ( - - ) 1255 Network ---------------------------------------------- 1257 Figure 4: Architecture of an IETF Network Slice 1259 The network itself (at the bottom of the figure) comprises an 1260 underlay network. This could be a physical network, but may be a 1261 virtual network. The underlay network is provisioned through network 1262 controllers that may utilize device controllers [RFC8309]. 1264 The underlay network may optionally be filtered or customized by the 1265 network operator to produce a number of network topologies that we 1266 call Filter Topologies. Customization is just a way of selecting 1267 specific resources (e.g., nodes and links) from the underlay network 1268 according to their capabilities and connectivity in the underlay 1269 network. These actions are configuration options or operator 1270 policies. The resulting topologies can be used as candidates to host 1271 IETF Network Slices and provide a useful way for the network operator 1272 to know in advance that all of the resources they are using to plan 1273 an IETF Network Slice would be able to meet specific SLOs and SLEs. 1274 The creation of a Filter Topology could be an offline planning 1275 activity or could be performed dynamically as new demands arise. The 1276 use of Filter Topologies is entirely optional in the architecture, 1277 and IETF Network Slices could be hosted directly on the underlay 1278 network. 1280 Recall that an IETF Network Slice is a service requested by / 1281 provided for the customer. The IETF Network Slice service is 1282 expressed in terms of one or more connectivity constructs. An 1283 implementation or operator is free to limit the number of 1284 connectivity constructs in a slice to exactly one. Each connectivity 1285 construct is associated within the IETF Network Slice service request 1286 with a set of SLOs and SLEs. The set of SLOs and SLEs does not need 1287 to be the same for every connectivity construct in the slice, but an 1288 implementation or operator is free to require that all connectivity 1289 constructs in a slice have the same set of SLOs and SLEs. 1291 One or more connectivity constructs from one or more slices are 1292 mapped to a set of network resources called a Network Resource 1293 Partition (NRP). A single connectivity construct is mapped to only 1294 one NRP (that is, the relationship is many to one). An NRP may be 1295 chosen to support a specific connectivity construct because of its 1296 ability to support a specific set of SLOs and SLEs, or its ability to 1297 support particular connectivity types, or for any administrative or 1298 operational reason. An implementation or operator is free to map 1299 each connectivity construct to a separate NRP, although there may be 1300 scaling implications depending on the solution implemented. Thus, 1301 the connectivity constructs from one slice may be mapped to one or 1302 more NRPs. By implication from the above, an implementation or 1303 operator is free to map all the connectivity constructs in a slice to 1304 a single NRP, and to not share that NRP with connectivity constructs 1305 from another slice. 1307 An NRP is simply a collection of resources identified in the underlay 1308 network. Thus, the NRP is a scoped view of a topology and may be 1309 considered as a topology in its own right. The process of 1310 determining the NRP may be made easier if the underlay network 1311 topology is first filtered into a Filter Topology in order to be 1312 aware of the subset of network resources that are suitable for 1313 specific NRPs, but this is optional. 1315 The steps described here can be applied in a variety of orders 1316 according to implementation and deployment preferences. Furthermore, 1317 the steps may be iterative so that the components are continually 1318 refined and modified as network conditions change and as service 1319 requests are received or relinquished, and even the underlay network 1320 could be extended if necessary to meet the customers' demands. 1322 6.2. Procedures to Realize IETF Network Slices 1324 There are a number of different technologies that can be used in the 1325 underlay, including physical connections, MPLS, time-sensitive 1326 networking (TSN), Flex-E, etc. 1328 An IETF Network Slice can be realized in a network, using specific 1329 underlay technology or technologies. The creation of a new IETF 1330 Network Slice will be realized with following steps: 1332 * The NSC exposes the network slicing capabilities that it offers 1333 for the network it manages. 1335 * The customer may issue a request to determine whether a specific 1336 IETF Network Slice could be supported by the network. The NSC may 1337 respond indicating a simple yes or no, and may supplement a 1338 negative response with information about what it could support 1339 were the customer to change some requirements. 1341 * The customer requests an IETF Network Slice. The NSC may respond 1342 that the slice has or has not been created, and may supplement a 1343 negative response with information about what it could support 1344 were the customer to change some requirements. 1346 * When processing a customer request for an IETF Network Slice, the 1347 NSC maps the request to the network capabilities and applies 1348 provider policies before creating or supplementing the NRP. 1350 Regardless of how IETF Network Slice is realized in the network 1351 (i.e., using tunnels of different types), the definition of the IETF 1352 Network Slice service does not change at all. The only difference is 1353 how the slice is realized. The following sections briefly introduce 1354 how some existing architectural approaches can be applied to realize 1355 IETF Network Slices. 1357 6.3. Applicability of ACTN to IETF Network Slices 1359 Abstraction and Control of TE Networks (ACTN - [RFC8453]) is a 1360 management architecture and toolkit used to create virtual networks 1361 (VNs) on top of a TE underlay network. The VNs can be presented to 1362 customers for them to operate as private networks. 1364 In many ways, the function of ACTN is similar to IETF network 1365 slicing. Customer requests for connectivity-based overlay services 1366 are mapped to dedicated or shared resources in the underlay network 1367 in a way that meets customer guarantees for service level objectives 1368 and for separation from other customers' traffic. [RFC8453] 1369 describes the function of ACTN as collecting resources to establish a 1370 logically dedicated virtual network over one or more TE networks. 1371 Thus, in the case of a TE-enabled underlay network, the ACTN VN can 1372 be used as a basis to realize IETF network slicing. 1374 While the ACTN framework is a generic VN framework that can be used 1375 for VN services beyond the IETF Network Slice, it also a suitable 1376 basis for delivering and realizing IETF Network Slices. 1378 Further discussion of the applicability of ACTN to IETF Network 1379 Slices including a discussion of the relevant YANG models can be 1380 found in [I-D.ietf-teas-applicability-actn-slicing]. 1382 6.4. Applicability of Enhanced VPNs to IETF Network Slices 1384 An enhanced VPN (VPN+) is designed to support the needs of new 1385 applications, particularly applications that are associated with 5G 1386 services, by utilizing an approach that is based on existing VPN and 1387 TE technologies and adds characteristics that specific services 1388 require over and above VPNs as they have previously been specified. 1390 An enhanced VPN can be used to provide enhanced connectivity services 1391 between customer sites and can be used to create the infrastructure 1392 to underpin a network slicing service. 1394 It is envisaged that enhanced VPNs will be delivered using a 1395 combination of existing, modified, and new networking technologies. 1397 [I-D.ietf-teas-enhanced-vpn] describes the framework for Enhanced 1398 Virtual Private Network (VPN+) services. 1400 6.5. Network Slicing and Aggregation in IP/MPLS Networks 1402 Network slicing provides the ability to partition a physical network 1403 into multiple isolated logical networks of varying sizes, structures, 1404 and functions so that each slice can be dedicated to specific 1405 services or customers. 1407 Many approaches are currently being worked on to support IETF Network 1408 Slices in IP and MPLS networks with or without the use of Segment 1409 Routing. Most of these approaches utilize a way of marking packets 1410 so that network nodes can apply specific routing and forwarding 1411 behaviors to packets that belong to different IETF Network Slices. 1412 Different mechanisms for marking packets have been proposed 1413 (including using MPLS labels and Segment Routing segment IDs) and 1414 those mechanisms are agnostic to the path control technology used 1415 within the underlay network. 1417 These approaches are also sensitive to the scaling concerns of 1418 supporting a large number of IETF Network Slices within a single IP 1419 or MPLS network, and so offer ways to aggregate the connectivity 1420 constructs of slices (or whole slices) so that the packet markings 1421 indicate an aggregate or grouping where all of the packets are 1422 subject to the same routing and forwarding behavior. 1424 At this stage, it is inappropriate to mention any of these proposed 1425 solutions that are currently work in progress and not yet adopted as 1426 IETF work. 1428 6.6. Network Slicing and Service Function Chaining (SFC) 1430 A customer may request an IETF Network Slice service that involves a 1431 set of service functions (SFs) together with the order in which these 1432 SFs are invoked. Also, the customer can specify the service 1433 objectives to be met by the underly network (e.g., one-way delay to 1434 cross a service function path, one-way delay to reach a specific SF). 1435 These SFs are considered as ancillary SDPs and are possibly 1436 placeholders (i.e., the SFs are identified, but not their locators). 1438 Service Function Chaining (SFC) [RFC7665] techniques can be used by a 1439 provider to instantiate such an IETF Network Service Slice. The NSC 1440 may proceed as follows. 1442 * Expose a set of ancillary SDPs that are hosted in the underlay 1443 network. 1445 * Capture the SFC requirements (including, traffic performance 1446 metrics) from the customer. One or more service chains may be 1447 associated with the same IETF Network Slice service as 1448 connectivity constructs. 1450 * Execute an SF placement algorithm to decide where to locate the 1451 ancillary SDPs in order to fulfil the service objectives. 1453 * Generate SFC classification rules to identify (part of) the slice 1454 traffic that will be bound to an SFC. These classification rules 1455 may be the same as or distinct from the identification rules used 1456 to bind incoming traffic to the associated IETF Network Slice. 1458 The NSC also generates a set of SFC forwarding policies that 1459 govern how the traffic will be forwarded along a service function 1460 path (SFP). 1462 * Identify the appropriate Classifiers in the underlay network and 1463 provision them with the classification rules. Likewise, the NSC 1464 communicates the SFC forwarding polices to the appropriate Service 1465 Function Forwarders (SFF). 1467 The provider can enable an SFC data plane mechanism, such as 1468 [RFC8300], [RFC8596], or [I-D.ietf-spring-nsh-sr]. 1470 7. Isolation in IETF Network Slices 1472 7.1. Isolation as a Service Requirement 1474 An IETF Network Slice customer may request that the IETF Network 1475 Slice delivered to them is such that changes to other IETF Network 1476 Slices or to other services do not have any negative impact on the 1477 delivery of the IETF Network Slice. The IETF Network Slice customer 1478 may specify the degree to which their IETF Network Slice is 1479 unaffected by changes in the provider network or by the behavior of 1480 other IETF Network Slice customers. The customer may express this 1481 via an SLE it agrees with the provider. This concept is termed 1482 'isolation'. 1484 In general, a customer cannot tell whether a service provider is 1485 meeting an isolation SLE. If the service varies such that an SLO is 1486 breached then the customer will become aware of the problem, and if 1487 the service varies within the allowed bounds of the SLOs, there may 1488 be no noticeable indication that this SLE has been violated. 1490 7.2. Isolation in IETF Network Slice Realization 1492 Isolation may be achieved in the underlay network by various forms of 1493 resource partitioning ranging from dedicated allocation of resources 1494 for a specific IETF Network Slice, to sharing of resources with 1495 safeguards. For example, traffic separation between different IETF 1496 Network Slices may be achieved using VPN technologies, such as L3VPN, 1497 L2VPN, EVPN, etc. Interference avoidance may be achieved by network 1498 capacity planning, allocating dedicated network resources, traffic 1499 policing or shaping, prioritizing in using shared network resources, 1500 etc. Finally, service continuity may be ensured by reserving backup 1501 paths for critical traffic, dedicating specific network resources for 1502 a selected number of IETF Network Slices. 1504 8. Management Considerations 1506 IETF Network Slice realization needs to be instrumented in order to 1507 track how it is working, and it might be necessary to modify the IETF 1508 Network Slice as requirements change. Dynamic reconfiguration might 1509 be needed. 1511 The various management interfaces and components are discussed in 1512 Section 5. 1514 9. Security Considerations 1516 This document specifies terminology and has no direct effect on the 1517 security of implementations or deployments. In this section, a few 1518 of the security aspects are identified. 1520 Conformance to security constraints: Specific security requests from 1521 customer-defined IETF Network Slices will be mapped to their 1522 realization in the underlay networks. Underlay networks will 1523 require capabilities to conform to customer's requests as some 1524 aspects of security may be expressed in SLEs. 1526 IETF NSC authentication: Underlay networks need to be protected 1527 against the attacks from an adversary NSC as this could 1528 destabilize overall network operations. An IETF Network Slice may 1529 span across different networks, therefore, the NSC should have 1530 strong authentication with each of these networks. Furthermore, 1531 both the IETF Network Slice Service Interface and the Network 1532 Configuration Interface need to be secured. 1534 Specific isolation criteria: The nature of conformance to isolation 1535 requests means that it should not be possible to attack an IETF 1536 Network Slice service by varying the traffic on other services or 1537 slices carried by the same underlay network. In general, 1538 isolation is expected to strengthen the IETF Network Slice 1539 security. 1541 Data Integrity of an IETF Network Slice: A customer wanting to 1542 secure their data and keep it private will be responsible for 1543 applying appropriate security measures to their traffic and not 1544 depending on the network operator that provides the IETF Network 1545 Slice. It is expected that for data integrity, a customer is 1546 responsible for end-to-end encryption of its own traffic. 1548 Note: See [NGMN_SEC] on 5G network slice security for discussion 1549 relevant to this section. 1551 IETF Network Slices might use underlying virtualized networking. All 1552 types of virtual networking require special consideration to be given 1553 to the separation of traffic between distinct virtual networks, as 1554 well as some degree of protection from effects of traffic use of 1555 underlay network (and other) resources from other virtual networks 1556 sharing those resources. 1558 For example, if a service requires a specific upper bound of latency, 1559 then that service can be degraded by added delay in transmission of 1560 service packets caused by the activities of another service or 1561 application using the same resources. 1563 Similarly, in a network with virtual functions, noticeably impeding 1564 access to a function used by another IETF Network Slice (for 1565 instance, compute resources) can be just as service-degrading as 1566 delaying physical transmission of associated packet in the network. 1568 While an IETF Network Slice might include encryption and other 1569 security features as part of the service, customers might be well 1570 advised to take responsibility for their own security needs, possibly 1571 by encrypting traffic before hand-off to a service provider. 1573 10. Privacy Considerations 1575 Privacy of IETF Network Slice service customers must be preserved. 1576 It should not be possible for one IETF Network Slice customer to 1577 discover the presence of other customers, nor should sites that are 1578 members of one IETF Network Slice be visible outside the context of 1579 that IETF Network Slice. 1581 In this sense, it is of paramount importance that the system use the 1582 privacy protection mechanism defined for the specific underlay 1583 technologies that support the slice, including in particular those 1584 mechanisms designed to preclude acquiring identifying information 1585 associated with any IETF Network Slice customer. 1587 11. IANA Considerations 1589 This document makes no requests for IANA action. 1591 12. Informative References 1593 [HIPAA] HHS, "Health Insurance Portability and Accountability Act 1594 - The Security Rule", February 2003, 1595 . 1598 [I-D.ietf-opsawg-sap] 1599 Boucadair, M., Dios, O. G. D., Barguil, S., Wu, Q., and V. 1600 Lopez, "A Network YANG Model for Service Attachment Points 1601 (SAPs)", Work in Progress, Internet-Draft, draft-ietf- 1602 opsawg-sap-02, 22 February 2022, 1603 . 1606 [I-D.ietf-spring-nsh-sr] 1607 Guichard, J. N. and J. Tantsura, "Integration of Network 1608 Service Header (NSH) and Segment Routing for Service 1609 Function Chaining (SFC)", Work in Progress, Internet- 1610 Draft, draft-ietf-spring-nsh-sr-10, 13 December 2021, 1611 . 1614 [I-D.ietf-teas-applicability-actn-slicing] 1615 King, D., Drake, J., Zheng, H., and A. Farrel, 1616 "Applicability of Abstraction and Control of Traffic 1617 Engineered Networks (ACTN) to Network Slicing", Work in 1618 Progress, Internet-Draft, draft-ietf-teas-applicability- 1619 actn-slicing-01, 7 March 2022, 1620 . 1623 [I-D.ietf-teas-enhanced-vpn] 1624 Dong, J., Bryant, S., Li, Z., Miyasaka, T., and Y. Lee, "A 1625 Framework for Enhanced Virtual Private Network (VPN+) 1626 Services", Work in Progress, Internet-Draft, draft-ietf- 1627 teas-enhanced-vpn-10, 6 March 2022, 1628 . 1631 [I-D.openconfig-rtgwg-gnmi-spec] 1632 Shakir, R., Shaikh, A., Borman, P., Hines, M., Lebsack, 1633 C., and C. Morrow, "gRPC Network Management Interface 1634 (gNMI)", Work in Progress, Internet-Draft, draft- 1635 openconfig-rtgwg-gnmi-spec-01, 5 March 2018, 1636 . 1639 [MACsec] IEEE, "IEEE Standard for Local and metropolitan area 1640 networks - Media Access Control (MAC) Security", 2018, 1641 . 1643 [NGMN-NS-Concept] 1644 NGMN Alliance, "Description of Network Slicing Concept", 1645 https://www.ngmn.org/uploads/ 1646 media/161010_NGMN_Network_Slicing_framework_v1.0.8.pdf , 1647 2016. 1649 [NGMN_SEC] NGMN Alliance, "NGMN 5G Security - Network Slicing", April 1650 2016, . 1653 [PCI] PCI Security Standards Council, "PCI DSS", May 2018, 1654 . 1656 [RFC3022] Srisuresh, P. and K. Egevang, "Traditional IP Network 1657 Address Translator (Traditional NAT)", RFC 3022, 1658 DOI 10.17487/RFC3022, January 2001, 1659 . 1661 [RFC3135] Border, J., Kojo, M., Griner, J., Montenegro, G., and Z. 1662 Shelby, "Performance Enhancing Proxies Intended to 1663 Mitigate Link-Related Degradations", RFC 3135, 1664 DOI 10.17487/RFC3135, June 2001, 1665 . 1667 [RFC3393] Demichelis, C. and P. Chimento, "IP Packet Delay Variation 1668 Metric for IP Performance Metrics (IPPM)", RFC 3393, 1669 DOI 10.17487/RFC3393, November 2002, 1670 . 1672 [RFC4208] Swallow, G., Drake, J., Ishimatsu, H., and Y. Rekhter, 1673 "Generalized Multiprotocol Label Switching (GMPLS) User- 1674 Network Interface (UNI): Resource ReserVation Protocol- 1675 Traffic Engineering (RSVP-TE) Support for the Overlay 1676 Model", RFC 4208, DOI 10.17487/RFC4208, October 2005, 1677 . 1679 [RFC4303] Kent, S., "IP Encapsulating Security Payload (ESP)", 1680 RFC 4303, DOI 10.17487/RFC4303, December 2005, 1681 . 1683 [RFC4364] Rosen, E. and Y. Rekhter, "BGP/MPLS IP Virtual Private 1684 Networks (VPNs)", RFC 4364, DOI 10.17487/RFC4364, February 1685 2006, . 1687 [RFC4397] Bryskin, I. and A. Farrel, "A Lexicography for the 1688 Interpretation of Generalized Multiprotocol Label 1689 Switching (GMPLS) Terminology within the Context of the 1690 ITU-T's Automatically Switched Optical Network (ASON) 1691 Architecture", RFC 4397, DOI 10.17487/RFC4397, February 1692 2006, . 1694 [RFC5212] Shiomoto, K., Papadimitriou, D., Le Roux, JL., Vigoureux, 1695 M., and D. Brungard, "Requirements for GMPLS-Based Multi- 1696 Region and Multi-Layer Networks (MRN/MLN)", RFC 5212, 1697 DOI 10.17487/RFC5212, July 2008, 1698 . 1700 [RFC5440] Vasseur, JP., Ed. and JL. Le Roux, Ed., "Path Computation 1701 Element (PCE) Communication Protocol (PCEP)", RFC 5440, 1702 DOI 10.17487/RFC5440, March 2009, 1703 . 1705 [RFC6020] Bjorklund, M., Ed., "YANG - A Data Modeling Language for 1706 the Network Configuration Protocol (NETCONF)", RFC 6020, 1707 DOI 10.17487/RFC6020, October 2010, 1708 . 1710 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 1711 NAT64: Network Address and Protocol Translation from IPv6 1712 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 1713 April 2011, . 1715 [RFC6241] Enns, R., Ed., Bjorklund, M., Ed., Schoenwaelder, J., Ed., 1716 and A. Bierman, Ed., "Network Configuration Protocol 1717 (NETCONF)", RFC 6241, DOI 10.17487/RFC6241, June 2011, 1718 . 1720 [RFC7665] Halpern, J., Ed. and C. Pignataro, Ed., "Service Function 1721 Chaining (SFC) Architecture", RFC 7665, 1722 DOI 10.17487/RFC7665, October 2015, 1723 . 1725 [RFC7679] Almes, G., Kalidindi, S., Zekauskas, M., and A. Morton, 1726 Ed., "A One-Way Delay Metric for IP Performance Metrics 1727 (IPPM)", STD 81, RFC 7679, DOI 10.17487/RFC7679, January 1728 2016, . 1730 [RFC7680] Almes, G., Kalidindi, S., Zekauskas, M., and A. Morton, 1731 Ed., "A One-Way Loss Metric for IP Performance Metrics 1732 (IPPM)", STD 82, RFC 7680, DOI 10.17487/RFC7680, January 1733 2016, . 1735 [RFC7926] Farrel, A., Ed., Drake, J., Bitar, N., Swallow, G., 1736 Ceccarelli, D., and X. Zhang, "Problem Statement and 1737 Architecture for Information Exchange between 1738 Interconnected Traffic-Engineered Networks", BCP 206, 1739 RFC 7926, DOI 10.17487/RFC7926, July 2016, 1740 . 1742 [RFC7950] Bjorklund, M., Ed., "The YANG 1.1 Data Modeling Language", 1743 RFC 7950, DOI 10.17487/RFC7950, August 2016, 1744 . 1746 [RFC8040] Bierman, A., Bjorklund, M., and K. Watsen, "RESTCONF 1747 Protocol", RFC 8040, DOI 10.17487/RFC8040, January 2017, 1748 . 1750 [RFC8300] Quinn, P., Ed., Elzur, U., Ed., and C. Pignataro, Ed., 1751 "Network Service Header (NSH)", RFC 8300, 1752 DOI 10.17487/RFC8300, January 2018, 1753 . 1755 [RFC8309] Wu, Q., Liu, W., and A. Farrel, "Service Models 1756 Explained", RFC 8309, DOI 10.17487/RFC8309, January 2018, 1757 . 1759 [RFC8453] Ceccarelli, D., Ed. and Y. Lee, Ed., "Framework for 1760 Abstraction and Control of TE Networks (ACTN)", RFC 8453, 1761 DOI 10.17487/RFC8453, August 2018, 1762 . 1764 [RFC8454] Lee, Y., Belotti, S., Dhody, D., Ceccarelli, D., and B. 1765 Yoon, "Information Model for Abstraction and Control of TE 1766 Networks (ACTN)", RFC 8454, DOI 10.17487/RFC8454, 1767 September 2018, . 1769 [RFC8596] Malis, A., Bryant, S., Halpern, J., and W. Henderickx, 1770 "MPLS Transport Encapsulation for the Service Function 1771 Chaining (SFC) Network Service Header (NSH)", RFC 8596, 1772 DOI 10.17487/RFC8596, June 2019, 1773 . 1775 [TS23501] 3GPP, "System architecture for the 5G System (5GS)", 1776 3GPP TS 23.501, 2019. 1778 [TS28530] 3GPP, "Management and orchestration; Concepts, use cases 1779 and requirements", 3GPP TS 28.530, 2019. 1781 [TS33.210] 3GPP, "3G security; Network Domain Security (NDS); IP 1782 network layer security (Release 14).", December 2016, 1783 . 1786 Acknowledgments 1788 The entire TEAS Network Slicing design team and everyone 1789 participating in related discussions has contributed to this 1790 document. Some text fragments in the document have been copied from 1791 the [I-D.ietf-teas-enhanced-vpn], for which we are grateful. 1793 Significant contributions to this document were gratefully received 1794 from the contributing authors listed in the "Contributors" section. 1795 In addition we would like to also thank those others who have 1796 attended one or more of the design team meetings, including the 1797 following people not listed elsewhere: 1799 * Aihua Guo 1801 * Bo Wu 1803 * Greg Mirsky 1805 * Lou Berger 1807 * Rakesh Gandhi 1809 * Ran Chen 1811 * Sergio Belotti 1813 * Stewart Bryant 1815 * Tomonobu Niwa 1816 * Xuesong Geng 1818 Further useful comments were received from Daniele Ceccarelli, Uma 1819 Chunduri, Pavan Beeram, Tarek Saad, Kenichi Ogaki, Oscar Gonzalez de 1820 Dios, Xiaobing Niu, Dan Voyer, Igor Bryskin, Luay Jalil, Joel 1821 Halpern, John Scudder, John Mullooly, and Krzysztof Szarkowicz. 1823 This work is partially supported by the European Commission under 1824 Horizon 2020 grant agreement number 101015857 Secured autonomic 1825 traffic management for a Tera of SDN flows (Teraflow). 1827 Contributors 1829 The following authors contributed significantly to this document: 1831 Eric Gray 1832 (The original editor of the foundation documents) 1833 Independent 1834 Email: ewgray@graiymage.com 1836 Jari Arkko 1837 Ericsson 1838 Email: jari.arkko@piuha.net 1840 Mohamed Boucadair 1841 Orange 1842 Email: mohamed.boucadair@orange.com 1844 Dhruv Dhody 1845 Huawei, India 1846 Email: dhruv.ietf@gmail.com 1848 Jie Dong 1849 Huawei 1850 Email: jie.dong@huawei.com 1852 Xufeng Liu 1853 Volta Networks 1854 Email: xufeng.liu.ietf@gmail.com 1856 Authors' Addresses 1858 Adrian Farrel (editor) 1859 Old Dog Consulting 1860 United Kingdom 1861 Email: adrian@olddog.co.uk 1862 John Drake (editor) 1863 Juniper Networks 1864 United States of America 1865 Email: jdrake@juniper.net 1867 Reza Rokui 1868 Ciena 1869 Email: rrokui@ciena.com 1871 Shunsuke Homma 1872 NTT 1873 Japan 1874 Email: shunsuke.homma.ietf@gmail.com 1876 Kiran Makhijani 1877 Futurewei 1878 United States of America 1879 Email: kiranm@futurewei.com 1881 Luis M. Contreras 1882 Telefonica 1883 Spain 1884 Email: luismiguel.contrerasmurillo@telefonica.com 1886 Jeff Tantsura 1887 Microsoft Inc. 1888 Email: jefftant.ietf@gmail.com