idnits 2.17.1 draft-ietf-ccamp-wavelength-switched-framework-01.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 20. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 1600. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 1577. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 1584. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 1590. 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 Copyright Line does not match the current year -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (October 31, 2008) is 5650 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'G.872' is defined on line 1470, but no explicit reference was found in the text == Outdated reference: A later version (-10) exists of draft-ietf-pce-global-concurrent-optimization-05 == Outdated reference: A later version (-19) exists of draft-ietf-pce-pcep-16 == Outdated reference: A later version (-06) exists of draft-ietf-pce-of-05 == Outdated reference: A later version (-01) exists of draft-bernstein-ccamp-wson-encode-00 Summary: 1 error (**), 0 flaws (~~), 6 warnings (==), 8 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 Network Working Group G. Bernstein (ed.) 2 Internet Draft Grotto Networking 3 Intended status: Informational Y. Lee (ed.) 4 Expires: April 2009 Huawei 5 Wataru Imajuku 6 NTT 8 October 31, 2008 10 Framework for GMPLS and PCE Control of Wavelength Switched Optical 11 Networks (WSON) 12 draft-ietf-ccamp-wavelength-switched-framework-01.txt 14 Status of this Memo 16 By submitting this Internet-Draft, each author represents that 17 any applicable patent or other IPR claims of which he or she is 18 aware have been or will be disclosed, and any of which he or she 19 becomes aware will be disclosed, in accordance with Section 6 of 20 BCP 79. 22 Internet-Drafts are working documents of the Internet Engineering 23 Task Force (IETF), its areas, and its working groups. Note that 24 other groups may also distribute working documents as Internet- 25 Drafts. 27 Internet-Drafts are draft documents valid for a maximum of six months 28 and may be updated, replaced, or obsoleted by other documents at any 29 time. It is inappropriate to use Internet-Drafts as reference 30 material or to cite them other than as "work in progress." 32 The list of current Internet-Drafts can be accessed at 33 http://www.ietf.org/ietf/1id-abstracts.txt 35 The list of Internet-Draft Shadow Directories can be accessed at 36 http://www.ietf.org/shadow.html 38 This Internet-Draft will expire on April 31, 2009. 40 Copyright Notice 42 Copyright (C) The IETF Trust (2008). 44 Abstract 45 This memo provides a framework for applying Generalized Multi- 46 Protocol Label Switching (GMPLS) and the Path Computation Element 47 (PCE) architecture to the control of wavelength switched optical 48 networks (WSON). In particular we provide control plane models for 49 key wavelength switched optical network subsystems and processes. The 50 subsystems include wavelength division multiplexed links, tunable 51 laser transmitters, reconfigurable optical add/drop multiplexers 52 (ROADM) and wavelength converters. 54 Lightpath provisioning, in general, requires the routing and 55 wavelength assignment (RWA) process. This process is reviewed and the 56 information requirements, both static and dynamic for this process 57 are presented, along with alternative implementation scenarios that 58 could be realized via GMPLS/PCE and/or extended GMPLS/PCE protocols. 59 This memo does NOT address optical impairments in any depth and 60 focuses on topological elements and path selection constraints that 61 are common across different WSON environments. It is expected that a 62 variety of different techniques will be applied to optical 63 impairments depending on the type of WSON, such as access, metro or 64 long haul. 66 Table of Contents 68 1. Introduction...................................................3 69 2. Terminology....................................................4 70 3. Wavelength Switched Optical Networks...........................5 71 3.1. WDM and CWDM Links........................................5 72 3.2. Optical Transmitters......................................7 73 3.2.1. Lasers...............................................7 74 3.2.2. Spectral Characteristics & Modulation Type...........8 75 3.2.3. Signal Rates and Error Correction....................9 76 3.3. ROADMs, OXCs, Splitters, Combiners and FOADMs............10 77 3.3.1. Reconfigurable Add/Drop Multiplexers and OXCs.......10 78 3.3.2. Splitters...........................................12 79 3.3.3. Combiners...........................................12 80 3.3.4. Fixed Optical Add/Drop Multiplexers.................12 81 3.4. Wavelength Converters....................................13 82 4. Routing and Wavelength Assignment and the Control Plane.......15 83 4.1. Architectural Approaches to RWA..........................16 84 4.1.1. Combined RWA (R&WA).................................16 85 4.1.2. Separated R and WA (R+WA)...........................17 86 4.1.3. Routing and Distributed WA (R+DWA)..................17 87 4.2. Conveying information needed by RWA......................18 88 4.3. Lightpath Temporal Characteristics.......................19 89 5. GMPLS & PCE Implications......................................20 90 5.1. Implications for GMPLS signaling.........................20 91 5.1.1. Identifying Wavelengths and Signals.................20 92 5.1.2. Combined RWA/Separate Routing WA support............20 93 5.1.3. Distributed Wavelength Assignment: Unidirectional, No 94 Converters.................................................21 95 5.1.4. Distributed Wavelength Assignment: Unidirectional, 96 Limited Converters.........................................22 97 5.1.5. Distributed Wavelength Assignment: Bidirectional, No 98 Converters.................................................22 99 5.2. Implications for GMPLS Routing...........................23 100 5.2.1. Need for Wavelength-Specific Maximum Bandwidth 101 Information................................................23 102 5.2.2. Need for Wavelength-Specific Availability Information24 103 5.2.3. Relationship to Link Bundling and Layering..........24 104 5.2.4. WSON Routing Information Summary....................24 105 5.3. Optical Path Computation and Implications for PCE........26 106 5.3.1. Lightpath Constraints and Characteristics...........26 107 5.3.2. Computation Architecture Implications...............27 108 5.3.3. Discovery of RWA Capable PCEs.......................27 109 5.4. Scaling Implications.....................................27 110 5.4.1. Routing.............................................28 111 5.4.2. Signaling...........................................28 112 5.4.3. Path computation....................................28 113 5.5. Summary of Impacts by RWA Architecture...................28 114 6. Security Considerations.......................................29 115 7. IANA Considerations...........................................29 116 8. Acknowledgments...............................................30 117 9. References....................................................31 118 9.1. Normative References.....................................31 119 9.2. Informative References...................................32 120 10. Contributors.................................................35 121 Author's Addresses...............................................35 122 Intellectual Property Statement..................................36 123 Disclaimer of Validity...........................................37 125 1. Introduction 127 From its beginning Generalized Multi-Protocol Label Switching (GMPLS) 128 was intended to control wavelength switched optical networks (WSON) 129 with the GMPLS architecture document [RFC3945] explicitly mentioning 130 both wavelength and waveband switching and equating wavelengths 131 (lambdas) with GMPLS labels. In addition a discussion of optical 132 impairments and other constraints on optical routing can be found in 133 [RFC4054]. However, optical technologies have advanced in ways that 134 make them significantly different from other circuit switched 135 technologies such as Time Division Multiplexing (TDM). Service 136 providers have already deployed many of these new optical 137 technologies such as ROADMs and tunable lasers and desire the same 138 automation and restoration capabilities that GMPLS has provided to 139 TDM and packet switched networks. Another important application of an 140 automated control plane such as GMPLS is the possibility to improve, 141 via recovery schemes, the availability of the network. One of the 142 key points of GMPLS based recovery schemes is the capability to 143 survive multiple failures while legacy protection mechanism such as 144 1+1 path protection can survive from a single failure. Moreover this 145 improved availability can be obtained using less network resources. 147 This document will focus on the unique properties of links, switches 148 and path selection constraints that occur in WSONs. Different WSONs 149 such as access, metro and long haul may apply different techniques 150 for dealing with optical impairments hence this document will NOT 151 address optical impairments in any depth, but instead focus on 152 properties that are common across a variety of WSONs. 154 This memo provides a framework for applying GMPLS and the Path 155 Computation Element (PCE) architecture to the control of WSONs. In 156 particular we provide control plane models for key wavelength 157 switched optical network subsystems and processes. The subsystems 158 include wavelength division multiplexed links, tunable laser 159 transmitters, reconfigurable optical add/drop multiplexers (ROADM) 160 and wavelength converters. 162 Lightpath provisioning, in general, requires the routing and 163 wavelength assignment (RWA) process. This process is reviewed and the 164 information requirements, both static and dynamic for this process 165 are presented, along with alternative implementation architectures 166 that could be realized via various combinations of extended GMPLS and 167 PCE protocols. 169 2. Terminology 171 CWDM: Coarse Wavelength Division Multiplexing. 173 DWDM: Dense Wavelength Division Multiplexing. 175 FOADM: Fixed Optical Add/Drop Multiplexer. 177 OXC: Optical cross connect. A symmetric optical switching element in 178 which a signal on any ingress port can reach any egress port. 180 ROADM: Reconfigurable Optical Add/Drop Multiplexer. An asymmetric 181 wavelength selective switching element featuring ingress and egress 182 line side ports as well as add/drop side ports. 184 RWA: Routing and Wavelength Assignment. 186 Wavelength Conversion/Converters: The process of converting an 187 information bearing optical signal centered at a given wavelength to 188 one with "equivalent" content centered at a different wavelength. 189 Wavelength conversion can be implemented via an optical-electronic- 190 optical (OEO) process or via a strictly optical process. 192 WDM: Wavelength Division Multiplexing. 194 Wavelength Switched Optical Networks (WSON): WDM based optical 195 networks in which switching is performed selectively based on the 196 center wavelength of an optical signal. 198 3. Wavelength Switched Optical Networks 200 WSONs come in a variety of shapes and sizes from continent spanning 201 long haul networks, to metropolitan networks, to residential access 202 networks. In all these cases we are concerned with those properties 203 that constrain the choice of wavelengths that can be used, i.e., 204 restrict the wavelength label set, impact the path selection process, 205 and limit the topological connectivity. In the following we examine 206 and model some major subsystems of a WSON with an emphasis on those 207 aspects that are of relevance to the control plane. In particular we 208 look at WDM links, Optical Transmitters, ROADMs, and Wavelength 209 Converters. 211 3.1. WDM and CWDM Links 213 WDM and CWDM links run over optical fibers, and optical fibers come 214 in a wide range of types that tend to be optimized for various 215 applications from access networks, metro, long haul, and submarine 216 links to name a few. ITU-T and IEC standards exist for various types 217 of fibers. For the purposes here we are concerned only with single 218 mode fibers (SMF). The following SMF fiber types are typically 219 encountered in optical networks: 221 ITU-T Standard | Common Name 222 ------------------------------------------------------------ 223 G.652 [G.652] | Standard SMF | 224 G.653 [G.653] | Dispersion shifted SMF | 225 G.654 [G.654] | Cut-off shifted SMF | 226 G.655 [G.655] | Non-zero dispersion shifted SMF | 227 G.656 [G.656] | Wideband non-zero dispersion shifted SMF | 228 ------------------------------------------------------------ 229 These fiber types are differentiated by their optical impairment 230 characteristics such as attenuation, chromatic dispersion, 231 polarization mode dispersion, four wave mixing, etc. Since these 232 effects can be dependent upon wavelength, channel spacing and input 233 power level, the net effect for our modeling purposes here is to 234 restrict the range of wavelengths that can be used. 236 Typically WDM links operate in one or more of the approximately 237 defined optical bands [G.Sup39]: 239 Band Range (nm) Common Name Raw Bandwidth (THz) 240 O-band 1260-1360 Original 17.5 241 E-band 1360-1460 Extended 15.1 242 S-band 1460-1530 Short 9.4 243 C-band 1530-1565 Conventional 4.4 244 L-band 1565-1625 Long 7.1 245 U-band 1625-1675 Ultra-long 5.5 247 Not all of a band may be usable, for example in many fibers that 248 support E-band there is significant attenuation due to a water 249 absorption peak at 1383nm. Hence we can have a discontinuous 250 acceptable wavelength range for a particular link. Also some systems 251 will utilize more than one band. This is particularly true for coarse 252 WDM (CWDM) systems. 254 [Editor's note: the previous text is primarily tutorial in nature and 255 maybe deleted or moved to an appendix in a future draft] 257 Current technology breaks up the bandwidth capacity of fibers into 258 distinct channels based on either wavelength or frequency. There are 259 two standards covering wavelengths and channel spacing. ITU-T 260 recommendation [G.694.1] describes a DWDM grid defined in terms of 261 frequency grids of 12.5GHz, 25GHz, 50GHz, 100GHz, and other multiples 262 of 100GHz around a 193.1THz center frequency. At the narrowest 263 channel spacing this provides less than 4800 channels across the O 264 through U bands. ITU-T recommendation [G.694.2] describes a CWDM grid 265 define in terms of wavelength increments of 20nm running from 1271nm 266 to 1611nm for 18 or so channels. The number of channels is 267 significantly smaller than the 32 bit GMPLS label space allocated to 268 lambda switching. A fixed mapping between the GMPLS label space and 269 these ITU-T WDM grids as proposed in [Otani] would not only allow a 270 common vocabulary to be used in signaling lightpaths but also in 271 describing WDM links, ROADM ports, and wavelength converters for the 272 purposes path selection. 274 With a tremendous existing base of fiber many WDM links are designed 275 to take advantage of particular fiber characteristics or to try to 276 avoid undesirable properties. For example dispersion shifted SMF 277 [G.653] was originally designed for good long distance performance in 278 single channel systems, however putting WDM over this type of fiber 279 requires much system engineering and a fairly limited range of 280 wavelengths. Hence for our basic, impairment unaware, modeling of a 281 WDM link we will need the following information: 283 o Wavelength range(s): Given a mapping between labels and the ITU-T 284 grids each range could be expressed in terms of a doublet 285 (lambda1, lambda2) or (freq1, freq1) where the lambdas or 286 frequencies can be represented by 32 bit integers. 288 o Channel spacing: currently there are about five channel spacings 289 used in DWDM systems 12.5GHz to 200GHz and one defined CWDM 290 spacing. 292 For a particular link this information is relatively static, i.e., 293 changes to these properties generally require hardware upgrades. Such 294 information could be used locally during wavelength assignment via 295 signaling, similar to label restrictions in MPLS or used by a PCE in 296 solving the combined routing and wavelength assignment problem. 298 3.2. Optical Transmitters 300 3.2.1. Lasers 302 WDM optical systems make use of laser transmitters utilizing 303 different wavelengths (frequencies). Some laser transmitters were and 304 are manufactured for a specific wavelength of operation, that is, the 305 manufactured frequency cannot be changed. First introduced to reduce 306 inventory costs, tunable optical laser transmitters are becoming 307 widely deployed in some systems [Coldren04], [Buus06]. This allows 308 flexibility in the wavelength used for optical transmission and aids 309 in the control of path selection. 311 Fundamental modeling parameters from the control plane perspective 312 optical transmitters are: 314 o Tunable: Is this transmitter tunable or fixed. 316 o Tuning range: This is the frequency or wavelength range over which 317 the laser can be tuned. With the fixed mapping of labels to 318 lambda's of [Otani] this can be expressed as a doublet (lambda1, 319 lambda2) or (freq1, freq2) where lambda1 and lambda2 or freq1 and 320 freq2 are the labels representing the lower and upper bounds in 321 wavelength or frequency. 323 o Tuning time: Tuning times highly depend on the technology used. 324 Thermal drift based tuning may take seconds to stabilize, whilst 325 electronic tuning might provide sub-ms tuning times. Depending on 326 the application this might be critical. For example, thermal drift 327 might not be applicable for fast protection applications. 329 o Spectral Characteristics and stability: The spectral shape of the 330 laser's emissions and its frequency stability put limits on 331 various properties of the overall WDM system. One relatively easy 332 to characterize constraint is the finest channel spacing on which 333 the transmitter can be used. 335 Note that ITU-T recommendations specify many other aspects of a 336 laser's such as spectral characteristics and stability. Many of these 337 parameters are key in designing WDM subsystems consisting of 338 transmitters, WDM links and receivers however they do not furnish 339 additional information that will influence label switched path (LSP) 340 provisioning in a properly designed system. 342 Also note that lasers transmitters as a component can degrade and 343 fail over time. This presents the possibility of the failure of a LSP 344 (lightpath) without either a node or link failure. Hence, additional 345 mechanisms may be necessary to detect and differentiate this failure 346 from the others, e.g., one doesn't not want to initiate mesh 347 restoration if the source transmitter has failed, since the laser 348 transmitter will still be failed on the alternate optical path. 350 3.2.2. Spectral Characteristics & Modulation Type 352 Contrary to some marketing claims optical systems are not truly 353 "transparent" to the content of the signals that they carry. Each 354 lightpath will have spectral characteristics based on its content, 355 and the spacing of wavelengths in a WDM link will ultimately put 356 constraints on that spectrum. 358 For analog signals such as used in closed access television (CATV) or 359 "radio over fiber" links spectral characteristics are given in terms 360 of various bandwidth measures. However digital signals consist of our 361 main focus here and in the ITU-T G series optical specifications. In 362 this case the spectral characteristics can be more accurately 363 inferred from the modulation format and the bit rate. 365 Although Non-Return to Zero (NRZ) is currently the dominant form of 366 optical modulation, new modulation formats are being researched 367 [Winzer06] and deployed. With a choice in modulation formats we no 368 longer have a one to one relationship between digital bandwidth in 369 bytes or bits per second and the amount of optical spectrum (optical 370 bandwidth) consumed. To simplify the specification of optical signals 371 the ITU-T, in recommendation G.959.1, combined a rate bound and 372 modulation format designator [G.959.1]. For example, two of the 373 signal classes defined in [G.959.1] are: 375 Optical tributary signal class NRZ 1.25G: 377 "Applies to continuous digital signals with non-return to zero line 378 coding, from nominally 622 Mbit/s to nominally 1.25 Gbit/s. Optical 379 tributary signal class NRZ 1.25G includes a signal with STM-4 bit 380 rate according to ITU-T Rec. G.707/Y.1322." Note that Gigabit 381 Ethernet falls into this signaling class as well. 383 Optical tributary signal class RZ 40G: 385 "Applies to continuous digital signals with return to zero line 386 coding, from nominally 9.9 Gbit/s to nominally 43.02 Gbit/s. 387 Optical tributary signal class RZ 40G includes a signal with STM- 388 256 bit rate according to ITU-T Rec. G.707/Y.1322 and OTU3 bit rate 389 according to ITU-T Rec. G.709/Y.1331." 391 From a modeling perspective we have: 393 o Analog signals: bandwidth parameters, e.g., 3dB parameters and 394 similar. 396 o Digital signals: there are predefined modulation bit rate classes 397 that we can encode. 399 This information can be important in constraining route selection, 400 for example some signals may not be compatible with some links or 401 wavelength converters. In addition it lets the endpoints understand 402 if it can process the signal. 404 3.2.3. Signal Rates and Error Correction 406 Although, the spectral characteristics of a signal determine its 407 basic compatibility with a WDM system, more information is generally 408 needed for various processing activities such as regeneration and 409 reception. Many digital signals such as Ethernet, G.709, and SDH have 410 well defined encoding which includes forward error correction (FEC). 411 However many subsystem vendors offer additional FEC options for a 412 given signal type. The use of different FECs can lead to different 413 overall signal rates. If the FEC and rate used is not compatible 414 between the sender and receiver the signal can not be correctly 415 processed. Note that the rates of "standard" signals may be extended 416 to accommodate different payloads. For example there are 417 transmitters capable of directly mapping 10GE LAN-PHY traffic into 418 G.709 ODU2 frame with slightly higher clock rate [G.Sup43]. 420 3.3. ROADMs, OXCs, Splitters, Combiners and FOADMs 422 Definitions of various optical devices and their parameters can be 423 found in [G.671], we only look at a subset of these and their non- 424 impairement related properties. 426 3.3.1. Reconfigurable Add/Drop Multiplexers and OXCs 428 Reconfigurable add/drop optical multiplexers (ROADM) have matured and 429 are available in different forms and technologies [Basch06]. This is 430 a key technology that allows wavelength based optical switching. A 431 classic degree-2 ROADM is shown in Figure 1. 433 Line side ingress +---------------------+ Line side egress 434 --->| |---> 435 | | 436 | ROADM | 437 | | 438 | | 439 +---------------------+ 440 | | | | o o o o 441 | | | | | | | | 442 O O O O | | | | 443 Tributary Side: Drop (egress) Add (ingress) 445 Figure 1 Degree-2 ROADM 447 The key feature across all ROADM types is their highly asymmetric 448 switching capability. In the ROADM of Figure 1, the "add" ingress 449 ports can only egress on the line side egress port and not on any of 450 the "drop" egress ports. The degree of a ROADM or switch is given by 451 the number of line side ports (ingress and egress) and does not 452 include the number of "add" or "drop" ports. Sometimes the "add" 453 "drop" ports are also called tributary ports. As the degree of the 454 ROADM increases beyond two it can have properties of both a switch 455 (OXC) and a multiplexer and hence we must know the switched 456 connectivity offered by such a network element to effectively utilize 457 it. A straight forward way to do this is via a "switched 458 connectivity" matrix A where Amn = 0 or 1, depending upon whether a 459 wavelength on ingress port m can be connected to egress port n 460 [Imajuku]. For the ROADM of Figure 1 the switched connectivity matrix 461 can be expressed as 462 Ingress Egress Port 463 Port #1 #2 #3 #4 #5 464 -------------- 465 #1: 1 1 1 1 1 466 #2 1 0 0 0 0 467 A = #3 1 0 0 0 0 468 #4 1 0 0 0 0 469 #5 1 0 0 0 0 471 Where ingress ports 2-5 are add ports, egress ports 2-5 are drop 472 ports and ingress port #1 and egress port #1 are the line side (WDM) 473 ports. 475 For ROADMs this matrix will be very sparse, and for OXCs the 476 complement of the matrix will be very sparse, compact encodings and 477 usage including high degree ROADMs/OXCs are given in [WSON-Encode]. 479 Additional constraints may also apply to the various ports in a 480 ROADM/OXC. In the literature of optical switches and ROADMs the 481 following restrictions/terms are used: 483 Colored port: An ingress or more typically an egress (drop) port 484 restricted to a single channel of fixed wavelength. 486 Colorless port: An ingress or more typically an egress (drop) port 487 restricted to a single channel of arbitrary wavelength. 489 In general a port on a ROADM could have any of the following 490 wavelength restrictions: 492 o Multiple wavelengths, full range port 494 o Single wavelength, full range port 496 o Single wavelength, fixed lambda port 498 o Multiple wavelengths, reduced range port (like wave band 499 switching) 501 To model these restrictions we need two pieces of information for 502 each port: (a) number of wavelengths, (b) wavelength range and 503 spacing. Note that this information is relatively static. More 504 complicated wavelength constraints are modeled in [WSON-Info]. 506 3.3.2. Splitters 508 An optical splitter consists of a single ingress port and two or more 509 egress ports. The ingress optical signaled is essentially copied 510 (with loss) to all egress ports. 512 Using the modeling notions of section 3.3.1. the ingress and egress 513 ports of a splitter would have the same wavelength restrictions. In 514 addition we can describe a splitter by a connectivity matrix Amn as 515 follows: 517 Ingress Egress Port 518 Port #1 #2 #3 ... #N 519 ----------------- 520 A = #1 1 1 1 ... 1 522 The difference from a simple ROADM is that this is not a switched 523 connectivity matrix but the fixed connectivity matrix of the device. 525 3.3.3. Combiners 527 A optical combiner is somewhat the dual of a splitter in that it has 528 a single multi-wavelength egress port and multiple ingress ports. 529 The contents of all the ingress ports are copied and combined to the 530 single egress port. The various ports may have different wavelength 531 restrictions. It is generally the responsibility of those using the 532 combiner to assure that wavelength collision does not occur on the 533 egress port. The fixed connectivity matrix Amn for a combiner would 534 look like: 536 Ingress Egress Port 537 Port #1 538 --- 539 #1: 1 540 #2 1 541 A = #3 1 542 ... 1 543 #N 1 545 3.3.4. Fixed Optical Add/Drop Multiplexers 547 A fixed optical add/drop multiplexer can alter the course of an 548 ingress wavelength in a preset way. In particular a particular 549 wavelength (or waveband) from a line side ingress port would be 550 dropped to a particular "tributary" egress port. Depending on the 551 device's fixed configuration that same wavelength may or may not be 552 "continued" to the line side egress port ("drop and continue" 553 operation). Further there may exist tributary ingress ports ("add" 554 ports) whose signals are combined with each other and "continued" 555 line side signals. 557 In general to represent the routing properties of an FOADM we need a 558 fixed connectivity matrix Amn as previously discussed and we need the 559 precise wavelength restrictions for all ingress and egress ports. 560 From the wavelength restrictions on the tributary egress ports (drop 561 ports) we can see what wavelengths have been dropped. From the 562 wavelength restrictions on the tributary ingress (add) ports we can 563 see which wavelengths have been added to the line side egress port. 564 Finally from the added wavelength information and the line side 565 egress wavelength restrictions we can infer which wavelengths have 566 been continued. 568 To summarize, the modeling methodology introduced in section 3.3.1. 569 consisting of a connectivity matrix and port wavelength restrictions 570 can be used to describe a large set of fixed optical devices such as 571 combiners, splitters and FOADMs. Hybrid devices consisting of both 572 switched and fixed parts are modeled in [WSON-Info]. 574 3.4. Wavelength Converters 576 Wavelength converters take an ingress optical signal at one 577 wavelength and emit an equivalent content optical signal at another 578 wavelength on egress. There are currently two approaches to building 579 wavelength converters. One approach is based on optical to electrical 580 to optical (OEO) conversion with tunable lasers on egress. This 581 approach can be dependent upon the signal rate and format, i.e., this 582 is basically an electrical regenerator combined with a tunable laser. 583 The other approach performs the wavelength conversion, optically via 584 non-linear optical effects, similar in spirit to the familiar 585 frequency mixing used in radio frequency systems, but significantly 586 harder to implement. Such processes/effects may place limits on the 587 range of achievable conversion. These may depend on the wavelength of 588 the input signal and the properties of the converter as opposed to 589 only the properties of the converter in the OEO case. Different WSON 590 system designs may choose to utilize this component to varying 591 degrees or not at all. 593 Current or envisioned contexts for wavelength converters are: 595 1. Wavelength conversion associated with OEO switches and tunable 596 laser transmitters. In this case there are plenty of converters to 597 go around since we can think of each tunable output laser 598 transmitter on an OEO switch as a potential wavelength converter. 600 2. Wavelength conversion associated with ROADMs/OXCs. In this case we 601 may have a limited amount of conversion available. Conversion could 602 be either all optical or via an OEO method. 604 3. Wavelength conversion associated with fixed devices such as FOADMs. 605 In this case we may have a limited amount of conversion. Also in 606 this case the conversion may be used as part of light path routing. 608 Based on the above contexts a tentative modeling approach for 609 wavelength converters could be as follows: 611 1. Wavelength converters can always be modeled as associated with 612 network elements. This includes fixed wavelength routing elements. 614 2. A network element may have full wavelength conversion capability, 615 i.e., any ingress port and wavelength, or a limited number of 616 wavelengths and ports. On a box with a limited number of 617 converters there also may exist restrictions on which ports can 618 reach the converters. Hence regardless of where the converters 619 actually are we can associate them with ingress ports. 621 3. Wavelength converters have range restrictions that are either 622 independent or dependent upon the ingress wavelength. [TBD: for 623 those that depend on ingress wavelength can we have a standard 624 formula? Also note that this type of converter introduces 625 additional optical impairments.] 627 4. Wavelength converters that are O-E-O based will have a restriction 628 based on the modulation format and transmission speed. 630 Note that since O-E-O wavelength converters also serve as 631 regenerators we can include regenerators in our model of wavelength 632 converters. O-E-O Regenerators come in three general types known as 633 1R, 2R, and 3R regenerators. 1R regenerators re-amplify the signal to 634 combat attenuation, 2R regenerators reshape as well as amplify the 635 signal, 3R regenerators amplify, reshape and retime the signal. As we 636 go from 1R to 3R regenerators the signal is ''cleaned up'' better but 637 at the same time the regeneration process becomes more dependent on 638 the signal characteristics such as format and rate. 640 In WSONs where wavelength converters are sparse we may actually see a 641 light path appear to loop or ''backtrack'' upon itself in order to 642 reach a wavelength converter prior to continuing on to its 643 destination. The lambda used on the "detour" out to the wavelength 644 converter would be different from that coming back from the "detour" 645 to the wavelength converter. 647 A model for an O-E-O wavelength converter would consist of: 649 o Input lambda or frequency range 651 o Output lambda or frequency range 653 o Equivalent regeneration level (1R, 2R, 3R) 655 o Signal restrictions if a 2R or 3R regeneration: formats and rates 657 [FFS: Model for an all optical wavelength converter] 659 4. Routing and Wavelength Assignment and the Control Plane 661 In wavelength switched optical networks consisting of tunable lasers 662 and wavelength selective switches with wavelength converters on every 663 interface, path selection is similar to the MPLS and TDM circuit 664 switched cases in that the labels, in this case wavelengths 665 (lambdas), have only local significance. That is, a wavelength- 666 convertible network with full wavelength-conversion capability at 667 each node is equivalent to a circuit-switched TDM network with full 668 time slot interchange capability; thus, the routing problem needs to 669 be addressed only at the level of the traffic engineered (TE) link 670 choice, and wavelength assignment can be resolved locally by the 671 switches on a hop-by-hop basis. 673 However, in the limiting case of an optical network with no 674 wavelength converters, a light path (optical channel - OCh -) needs a 675 route from source to destination and must pick a single wavelength 676 that can be used along that path without "colliding" with the 677 wavelength used by any other light path that may share an optical 678 fiber. This is sometimes referred to as a "wavelength continuity 679 constraint". To ease up on this constraint while keeping network 680 costs in check a limited number of wavelength converters maybe 681 introduce at key points in the network [Chu03]. 683 In the general case of limited or no wavelength converters this 684 computation is known as the Routing and Wavelength Assignment (RWA) 685 problem [HZang00]. The "hardness" of this problem is well documented. 686 There, however, exist a number of reasonable approximate methods for 687 its solution [HZang00]. 689 The inputs to the basic RWA problem are the requested light paths 690 source and destination, the networks topology, the locations and 691 capabilities of any wavelength converters, and the wavelengths 692 available on each optical link. The output from an algorithm solving 693 the RWA problem is an explicit route through ROADMs, a wavelength for 694 the optical transmitter, and a set of locations (generally associated 695 with ROADMs or switches) where wavelength conversion is to occur and 696 the new wavelength to be used on each component link after that point 697 in the route. 699 It is to be noted that choice of specific RWA algorithm is out of the 700 scope for this document. However there are a number of different 701 approaches to dealing with the RWA algorithm that can affect the 702 division of effort between signaling, routing and PCE. 704 4.1. Architectural Approaches to RWA 706 Two general computational approaches are taken to solving the RWA 707 problem some algorithms utilize a two step procedure of path 708 selection followed by wavelength assignment, and others solve the 709 problem in a combined fashion. 711 In the following, three different ways of performing RWA in 712 conjunction with the control plane are considered. The choice of one 713 of these architectural approaches over another generally impacts the 714 demands placed on the various control plane protocols. 716 4.1.1. Combined RWA (R&WA) 718 In this case, a unique entity is in charge of performing routing and 719 wavelength assignment. This choice assumes that computational entity 720 has sufficient WSON network link/nodal information and topology to be 721 able to compute RWA. This solution relies on a sufficient knowledge 722 of network topology, of available network resources and of network 723 nodes capabilities. This knowledge has to be accessible to the entity 724 performing the routing and wavelength assignment. 726 This solution is compatible with most known RWA algorithms, and in 727 particular those concerned with network optimization. On the other 728 hand, this solution requires up-to-date and detailed network 729 information dissemination. 731 Such a computational entity could reside in two different logical 732 places: 734 o In a separate Path Computation Element (PCE) which hence owns the 735 complete and updated knowledge of network state and provides path 736 computation services to node. 738 o In the Ingress node, in that case all nodes have the R&WA 739 functionality; the knowledge of the network state is obtained by a 740 periodic flooding of information provided by the other nodes. 742 4.1.2. Separated R and WA (R+WA) 744 In this case a first entity performs routing, while a second performs 745 wavelength assignment. The first entity furnishes one or more paths 746 to the second entity that will perform wavelength assignment and 747 possibly final path selection. 749 As the entities computing the path and the wavelength assignment are 750 separated, this constrains the class of RWA algorithms that may be 751 implemented. Although it may seem that algorithms optimizing a joint 752 usage of the physical and spectral paths are excluded from this 753 solution, many practical optimization algorithms only consider a 754 limited set of possible paths, e.g., as computed via a k-shortest 755 path algorithm [Ozdaglar03]. Hence although there is no guarantee 756 that the selected final route and wavelength offers the optimal 757 solution by allowing multiple routes to pass to the wavelength 758 selection process reasonable optimization can be performed. 760 The entity performing the routing assignment needs the topology 761 information of the network, whereas the entity performing the 762 wavelength assignment needs information on the network available 763 resources and on network nodes capabilities. 765 4.1.3. Routing and Distributed WA (R+DWA) 767 In this case a first entity performs routing, while wavelength 768 assignment is performed on a hop-by-hop manner along the previously 769 computed route. This mechanism relies on updating of a list of 770 potential wavelengths used to ensure the wavelength continuity 771 constraint. 773 As currently specified, the GMPLS protocol suite signaling protocol 774 can accommodate such an approach. Per [RFC3471], the Label Set 775 selection works according to an AND scheme. Each hop restricts the 776 Label Set sent to the next hop from the one received from the 777 previous hop by performing an AND operation between the wavelength 778 referred by the labels it includes with the one available on the 779 ongoing interface. The constraint to perform this AND operation is up 780 to the node local policy (even if one expects a consistent policy 781 configuration throughout a given transparency domain). When 782 wavelength conversion is performed at an intermediate node, a new 783 Label Set is generated. The egress nodes selects one label in the 784 Label Set received at the node, which is also up to the node local 785 policy. 787 Depending on these policies a spectral assignment may not be found or 788 one consuming too many conversion resources relatively to what a 789 dedicated wavelength assignment policy would have achieved. Hence, 790 this may generate higher blocking probabilities in a heavily loaded 791 network. 793 On the one hand, this solution may be empowered with some signaling 794 extensions to ease its functioning and possibly enhance its 795 performances relatively to blocking. On the other hand this solution 796 is not stressing the information dissemination processes. 798 The first entity may be a PCE or the ingress node of the LSP. This 799 solution is applicable inside network where resource optimization is 800 not the most crucial constraint. 802 4.2. Conveying information needed by RWA 804 The previous sections have characterized WSONs and lightpath 805 requests. In particular high level models of the information by the 806 RWA process were presented. We can view this information as either 807 static, changing with hardware changes (including possibly failures), 808 or dynamic, can change with subsequent lightpath provisioning. The 809 timeliness in which an entity involved in the RWA process is notified 810 of such changes is fairly situational. For example, for network 811 restoration purposes, learning of a hardware failure or of new 812 hardware coming online to provide restoration capability can be 813 critical. 814 Currently there are various methods for communicating RWA relevant 815 information, these include, but are not limited to: 817 o Existing control plane protocols such as GMPLS routing and 818 signaling. Note that routing protocols can be used to convey both 819 static and dynamic information. Static information currently 820 conveyed includes items like router options and such. 822 o Management protocols such as NetConf, SNMPv3, CLI, CORBA, or 823 others. 825 o Directory services and accompanying protocols. These are good for 826 the dissemination of relatively static information. Not intended 827 for dynamic information. 829 o Other techniques for dynamic information: messaging straight from 830 NEs to PCE to avoid flooding. This would be useful if the number 831 of PCEs is significantly less than number of WSON NEs. Or other 832 ways to limit flooding to "interested" NEs. 834 Mechanisms to improve scaling of dynamic information: 836 o Tailor message content to WSON. For example the use of wavelength 837 ranges, or wavelength occupation bit maps. 839 Utilize incremental updates if feasible. 841 4.3. Lightpath Temporal Characteristics 843 The temporal characteristics of a light path connection is another 844 aspect that can affect the choice of solution to the RWA process. For 845 our purposes here we look at the timeliness of connection 846 establishment/teardown, and the duration of the connection. 848 Connection Establishment/Teardown Timeliness can be thought of in 849 approximately three time frames: 851 1. Time Critical: For example those lightpath establishments used for 852 restoration of service or other high priority real time service 853 requests. 855 2. Soft time bounds: This is a more typical new connection request. 856 While expected to be responsive, there should be more time to take 857 into account network optimization. 859 3. Scheduled or Advanced reservations. Here lightpath connections are 860 requested significantly ahead of their intended "in service" time. 861 There is the potential for significant network optimization if 862 multiple lightpaths can be computed concurrently to achieve network 863 optimization objectives. 865 Lightpath connection duration has typically been thought of as 866 approximately three time frames: 868 1. Dynamic: those lightpaths with relatively short duration (holding 869 times). 871 2. Pseudo-static: lightpaths with moderately long durations. 873 3. Static: lightpaths with long durations. 875 Different types of RWA algorithms have been developed for dealing 876 with dynamic versus pseudo-static conditions. These can address 877 service provider's needs for: (a) network optimization, (b) 878 restoration, and (c) highly dynamic lightpath provisioning. 880 Hence we can model timescale related lightpath requirements via the 881 following notions: 883 o Batch or Sequential light path connection requests 884 o Timeliness of Connection establishment 886 o Duration of lightpath connection 888 5. GMPLS & PCE Implications 890 The presence and amount of wavelength conversion available at a 891 wavelength switching interface has an impact on the information that 892 needs to be transferred by the control plane (GMPLS) and the PCE 893 architecture. Current GMPLS and PCE standards can address the full 894 wavelength conversion case so the following will only address the 895 limited and no wavelength conversion cases. 897 5.1. Implications for GMPLS signaling 899 Basic support for WSON signaling already exists in GMPLS with the 900 lambda (value 9) LSP encoding type [RFC3471], or for G.709 compatible 901 optical channels, the LSP encoding type (value = 13) "G.709 Optical 902 Channel" from [RFC4328]. However a number of practical issues arise 903 in the identification of wavelengths and signals, and distributed 904 wavelength assignment processes which are discussed below. 906 5.1.1. Identifying Wavelengths and Signals 908 As previously stated a global fixed mapping between wavelengths and 909 labels simplifies the characterization of WDM links and WSON devices. 910 Furthermore such a mapping as described in [Otani] eases 911 communication between PCE and WSON PCCs. 913 An alternative to a global network map of labels to wavelengths would 914 be to use LMP to assign the map for each link then convey that 915 information to any path computation entities, e.g., label switch 916 routers or stand alone PCEs. The local label map approach will 917 require the label-set contents in the RSVP-TE Path message to be 918 translated every time the map changes between an incoming link and 919 the outgoing link. 921 In the future, it maybe worthwhile to define traffic parameters for 922 lambda LSPs that include a signal type field that includes modulation 923 format/rate information. This is similar to what was done in 924 reference [RFC4606] for SONET/SDH signal types. 926 5.1.2. Combined RWA/Separate Routing WA support 928 In either the combined RWA or separate routing WA cases, the node 929 initiating the signaling will have a route from the source to 930 destination along with the wavelengths (generalized labels) to be 931 used along portions of the path. Current GMPLS signaling supports an 932 explicit route object (ERO) and within an ERO an ERO Label subobject 933 can be use to indicate the wavelength to be used at a particular 934 node. In case the local label map approach is used the label sub- 935 object entry in the ERO has to be translated appropriately. 937 5.1.3. Distributed Wavelength Assignment: Unidirectional, No 938 Converters 940 GMPLS signaling for a uni-directional lightpath LSP allows for the 941 use of a label set object in the RSVP-TE path message. The processing 942 of the label set object to take the intersection of available lambdas 943 along a path can be performed resulting in the set of available 944 lambda being known to the destination that can then use a wavelength 945 selection algorithm to choose a lambda. For example, the following is 946 a non-exhaustive subset of wavelength assignment (WA) approaches 947 discussed in [HZang00]: 949 1. Random: Looks at all available wavelengths for the light path then 950 chooses from those available at random. 952 2. First Fit: Wavelengths are ordered, first available (on all links) 953 is chosen. 955 3. Most Used: Out of the wavelengths available on the path attempts 956 to select most use wavelength in network. 958 4. Least Loaded: For multi-fiber networks. Chooses the wavelength j 959 that maximizes minimum of the difference between the number of 960 fibers on link l and the number of fibers on link l with 961 wavelength j occupied. 963 As can be seen from the above short list, wavelength assignment 964 methods have differing information or processing requirements. The 965 information requirements of these methods are as follows: 967 1. Random: nothing more than the available wavelength set. 969 2. First Fit: nothing more than the available wavelength set. 971 3. Most Used: the available wavelength set and information on global 972 wavelength use in the network. 974 4. Least Loaded: the available wavelength set and information 975 concerning the wavelength dependent loading for each link (this 976 applies to multi-fiber links). This could be obtained via global 977 information or via supplemental information passed via the 978 signaling protocol. 980 In case (3) above the global information needed by the wavelength 981 assignment could be derived from suitably enhanced GMPLS routing. 982 Note however this information need not be accurate enough for 983 combined RWA computation. Currently, GMPLS signaling does not provide 984 a way to indicate that a particular wavelength assignment algorithm 985 should be used. 987 5.1.4. Distributed Wavelength Assignment: Unidirectional, Limited 988 Converters 990 The previous outlined the case with no wavelength converters. In the 991 case of wavelength converters, nodes with wavelength converters would 992 need to make the decision as to whether to perform conversion. One 993 indicator for this would be that the set of available wavelengths 994 which is obtained via the intersection of the incoming label set and 995 the egress links available wavelengths is either null or deemed too 996 small to permit successful completion. 998 At this point the node would need to remember that it will apply 999 wavelength conversion and will be responsible for assigning the 1000 wavelength on the previous lambda-contiguous segment when the RSVP-TE 1001 RESV message passes by. The node will pass on an enlarged label set 1002 reflecting only the limitations of the wavelength converter and the 1003 egress link. The record route option in RVSP-TE signaling can be used 1004 to show where wavelength conversion has taken place. 1006 5.1.5. Distributed Wavelength Assignment: Bidirectional, No 1007 Converters 1009 There are potential issues in the case of a bi-directional lightpath 1010 which requires the use of the same lambda in both directions. We can 1011 try to use the above procedure to determine the available 1012 bidirectional lambda set if we use the interpretation that the 1013 available label set is available in both directions. However, a 1014 problem, arises in that bidirectional LSPs setup, according to 1015 [RFC3471] section 4.1, is indicated by the presence of an upstream 1016 label in the path message. 1018 However, until the intersection of the available label sets is 1019 obtained, e.g., at the destination node and the wavelength assignment 1020 algorithm has been run the upstream label information will not be 1021 available. Hence currently distributed wavelength assignment with 1022 bidirectional lightpaths is not supported. 1024 5.2. Implications for GMPLS Routing 1026 GMPLS routing [RFC4202] currently defines an interface capability 1027 descriptor for "lambda switch capable" (LSC) which we can use to 1028 describe the interfaces on a ROADM or other type of wavelength 1029 selective switch. In addition to the topology information typically 1030 conveyed via an IGP, we would need to convey the following subsystem 1031 properties to minimally characterize a WSON: 1033 1. WDM Link properties (allowed wavelengths). 1035 2. Laser Transmitters (wavelength range). 1037 3. ROADM/FOADM properties (connectivity matrix, port wavelength 1038 restrictions). 1040 4. Wavelength Converter properties (per network element, may change if 1041 a common limited shared pool is used). 1043 In most cases we should be able to combine items (1) and (2) into the 1044 information in item (3). Except for the number of wavelength 1045 converters that are available in a shared pool, and the previous 1046 information is fairly static. In the next two sections we discuss 1047 dynamic available link bandwidth information. 1049 5.2.1. Need for Wavelength-Specific Maximum Bandwidth Information 1051 Difficulties are encountered when trying to use the bandwidth 1052 accounting methods of [RFC4202] and [RFC3630] to describe the 1053 availability of wavelengths on a WDM link. The current RFCs give 1054 three link resource measures: Maximum Bandwidth, Maximum Reservable 1055 Bandwidth, and Unreserved Bandwidth. Although these can be used to 1056 describe a WDM span they do not provide the fundamental information 1057 needed for RWA. We are not given the maximum bandwidth per wavelength 1058 for the span. If we did then we could use the aforementioned measures 1059 to tell us the maximum wavelength count and the number of available 1060 wavelengths. 1062 For example, suppose we have a 32 channel WDM span, and that the 1063 system in general supports ITU-T NRZ signals up to NRZ 10Gbps. 1064 Further suppose that the first 20 channels are carrying 1Gbps 1065 Ethernet, then the maximum bandwidth would be 320Gbps and the maximum 1066 reservable bandwidth would be 120Gbps (12 wavelengths). 1067 Alternatively, consider the case where the first 8 channels are 1068 carrying 2.5Gbps SDH STM-16 channels, then the maximum bandwidth 1069 would still be 320Gbps and the maximum reservable bandwidth would be 1070 240Gbps (24 wavelengths). 1072 Such information would be useful in the routing with distributed WA 1073 approach of section 4.1.3. 1075 5.2.2. Need for Wavelength-Specific Availability Information 1077 Even if we know the number of available wavelengths on a link, we 1078 actually need to know which specific wavelengths are available and 1079 which are occupied if we are going to run a combined RWA process or 1080 separate WA process as discussed in sections 4.1.1. 4.1.2. This is 1081 currently not possible with GMPLS routing extensions. 1083 In the routing extensions for GMPLS [RFC4202], requirements for 1084 layer-specific TE attributes are discussed. The RWA problem for 1085 optical networks without wavelength converters imposes an additional 1086 requirement for the lambda (or optical channel) layer: that of 1087 knowing which specific wavelengths are in use. Note that current 1088 dense WDM (DWDM) systems range from 16 channels to 128 channels with 1089 advanced laboratory systems with as many as 300 channels. Given these 1090 channel limitations and if we take the approach of a global 1091 wavelength to label mapping or furnishing the local mappings to the 1092 PCEs then representing the use of wavelengths via a simple bit-map is 1093 feasible. 1095 5.2.3. Relationship to Link Bundling and Layering 1097 When dealing with static DWDM systems, particularly from a SONET/SDH 1098 or G.709 digital wrapper layer, each lambda looks like a separate 1099 link. Typically a bunch of unnumbered links, as supported in GMPLS 1100 routing extensions [RFC4202], would be used to describe a static DWDM 1101 system. In addition these links can be bundled into a TE link 1102 ([RFC4202], [RFC4201]) for more efficient dissemination of resource 1103 information. However, in the case discussed here we want to control a 1104 dynamic WDM layer and must deal with wavelengths as labels and not 1105 just as links or component links from the perspective of an upper 1106 (client) layer. In addition, a typical point to point optical cable 1107 contains many optical fibers and hence it may be desirable to bundle 1108 these separate fibers into a TE link. Note that in the no wavelength 1109 conversion or limited wavelength conversion situations that we will 1110 need information on wavelength usage on the individual component 1111 links. 1113 5.2.4. WSON Routing Information Summary 1115 The following table summarizes the WSON information that could be 1116 conveyed via GMPLS routing and attempts to classify that information 1117 as to its static or dynamic nature and whether that information would 1118 tend to be associated with either a link or a node. 1120 Information Static/Dynamic Node/Link 1121 ------------------------------------------------------------------ 1122 Connectivity matrix Static Node 1123 Per port wavelength restrictions Static Node(1) 1124 WDM link (fiber) lambda ranges Static Link 1125 WDM link channel spacing Static Link 1126 Laser Transmitter range Static Link(2) 1127 Wavelength conversion capabilities Static(3) Node 1128 Maximum bandwidth per Wavelength Static Link 1129 Wavelength Availability Dynamic(4) Link 1131 Notes: 1133 1. These are the per port wavelength restrictions of an optical 1134 device such as a ROADM and are independent of any optical 1135 constraints imposed by a fiber link. 1137 2. This could also be viewed as a node capability. 1139 3. This could be dynamic in the case of a limited pool of converters 1140 where the number available can change with connection 1141 establishment. Note we may want to include regeneration 1142 capabilities here since OEO converters are also regenerators. 1144 4. Not necessarily needed in the case of distributed wavelength 1145 assignment via signaling. 1147 While the full complement of the information from the previous table 1148 is needed in the Combined RWA and the separate Routing and WA 1149 architectures, in the case of Routing + distribute WA via signaling 1150 we only need the following information: 1152 Information Static/Dynamic Node/Link 1153 ------------------------------------------------------------------ 1154 Connectivity matrix Static Node 1155 Wavelength conversion capabilities Static(3) Node 1157 Information models and compact encodings for this information is 1158 provided in [WSON-Info]. 1160 5.3. Optical Path Computation and Implications for PCE 1162 As previously noted the RWA problem can be computationally intensive 1163 [HZang00]. Such computationally intensive path computations and 1164 optimizations were part of the impetus for the PCE (path computation 1165 element) architecture. 1167 As the PCEP defines the procedures necessary to support both 1168 sequential [PCEP] and global concurrent path computations [PCE-GCO], 1169 PCE is well positioned to support WSON-enabled RWA computation with 1170 some protocol enhancement. 1172 Implications for PCE generally fall into two main categories: (a) 1173 lightpath constraints and characteristics, (b) computation 1174 architectures. 1176 5.3.1. Lightpath Constraints and Characteristics 1178 For the varying degrees of optimization that may be encountered in a 1179 network the following models of bulk and sequential lightpath 1180 requests are encountered: 1182 o Batch optimization, multiple lightpaths requested at one time. 1184 o Lightpath(s) and backup lightpath(s) requested at one time. 1186 o Single lightpath requested at a time. 1188 PCEP and PCE-GCO can be readily enhanced to support all of the 1189 potential models of RWA computation. 1191 Lightpath constraints include: 1193 o Bidirectional Assignment of wavelengths 1195 o Possible simultaneous assignment of wavelength to primary and 1196 backup paths. 1198 o Tuning range constraint on optical transmitter. 1200 Lightpath characteristics can include: 1202 o Duration information (how long this connection may last) 1204 o Timeliness/Urgency information (how quickly is this connection 1205 needed) 1206 5.3.2. Computation Architecture Implications 1208 When a PCE performs a combined RWA computation per section 4.1.1. it 1209 requires accurate an up to date wavelength utilization on all links 1210 in the network. 1212 When a PCE is used to perform wavelength assignment (WA) in the 1213 separate routing WA architecture then the entity requesting WA needs 1214 to furnish the pre-selected route to the PCE as well as any of the 1215 lightpath constraints/characteristics previously mentioned. This 1216 architecture also requires the PCE performing WA to have accurate and 1217 up to date network wavelength utilization information. 1219 When a PCE is used to perform routing in a routing with distribute WA 1220 architecture, then the PCE does not necessarily need the most up to 1221 date network wavelength utilization information, however timely 1222 information can contributed to reducing failed signaling attempts 1223 related to blocking. 1225 5.3.3. Discovery of RWA Capable PCEs 1227 The algorithms and network information needed for solving the RWA are 1228 somewhat specialized and computationally intensive hence not all PCEs 1229 within a domain would necessarily need or want this capability. 1230 Hence, it would be useful via the mechanisms being established for 1231 PCE discovery [RFC5088] to indicate that a PCE has the ability to 1232 deal with the RWA problem. Reference [RFC5088] indicates that a sub- 1233 TLV could be allocated for this purpose. 1235 Recent progress on objective functions in PCE [PCE-OF] would allow 1236 the operators to flexibly request differing objective functions per 1237 their need and applications. For instance, this would allow the 1238 operator to choose an objective function that minimizes the total 1239 network cost associated with setting up a set of paths concurrently. 1240 This would also allow operators to choose an objective function that 1241 results in a most evenly distributed link utilization. 1243 This implies that PCEP would easily accommodate wavelength selection 1244 algorithm in its objective function to be able to optimize the path 1245 computation from the perspective of wavelength assignment if chosen 1246 by the operators. 1248 5.4. Scaling Implications 1250 This section provides a summary of the scaling issue for WSON 1251 routing, signaling and path computation introduced by the concepts 1252 discussed in this document. 1254 5.4.1. Routing 1256 In large WSONs label availability and cross connect capability 1257 information being advertised may generate a significant amount of 1258 routing information. 1260 5.4.2. Signaling 1262 When dealing with a large number of simultaneous end-to-end 1263 wavelength service requests and service deletions the network may 1264 have to process a significant number of forward and backward service 1265 messages. Also, similar situation possibly happens in the case of 1266 link or node failure, if the WSON support dynamic restoration 1267 capability. 1269 5.4.3. Path computation 1271 If a PCE is handling path computation requests for end-to-end 1272 wavelength services within the WSON, then the complexity of the 1273 network and number of service path computation requests being sent to 1274 the PCE may have an impact on the PCEs ability to process requests in 1275 a timely manner. 1277 5.5. Summary of Impacts by RWA Architecture 1279 The following table summarizes for each RWA strategy the list of 1280 mandatory ("M") and optional ("O") control plane features according 1281 to GMPLS architectural blocks: 1283 o Information required by the path computation entity, 1285 o LSP request parameters used in either PCC to PCE situations or in 1286 signaling, 1288 o RSVP-TE LSP signaling parameters used in LSP establishment. 1290 The table shows which enhancements are common to all architectures 1291 (R&WA, R+WA, R+DWA), which apply only to R&WA and R+WA (R+&WA), and 1292 which apply only to R+DWA. 1294 +-------------------------------------+-----+-------+-------+-------+ 1295 | | |Common | R+&WA | R+DWA | 1296 | Feature | ref +---+---+---+---+---+---+ 1297 | | | M | O | M | O | M | O | 1298 +-------------------------------------+-----+---+---+---+---+---+---+ 1299 | Generalized Label for Wavelength |5.1.1| x | | | | | | 1300 +-------------------------------------+-----+---+---+---+---+---+---+ 1301 | Flooding of information for the | | | | | | | | 1302 | routing phase | | | | | | | | 1303 | Node features | 3.3 | | | | | | | 1304 | Node type | | | x | | | | | 1305 | spectral X-connect constraint | | | | x | | | | 1306 | port X-connect constraint | | | | x | | | | 1307 | Transponders availability | | | x | | | | | 1308 | Transponders features | 3.2 | | x | | | | | 1309 | Converter availability | | | | x | | | | 1310 | Converter features | 3.4 | | | x | | | x | 1311 | TE-parameters of WDM links | 3.1 | x | | | | | | 1312 | Total Number of wavelength | | x | | | | | | 1313 | Number of wavelengths available | | x | | | | | | 1314 | Grid spacing | | x | | | | | | 1315 | Wavelength availability on links | 5.2 | | | x | | | | 1316 +-------------------------------------+-----+---+---+---+---+---+---+ 1317 | LSP request parameters | | | | | | | | 1318 | Signal features | 5.1 | | x | | | x | | 1319 | Modulation format | | | x | | | x | | 1320 | Modulation parameters | | | x | | | x | | 1321 | Specification of RWA method | 5.1 | | x | | | x | | 1322 | LSP time features | 4.3 | | x | | | | | 1323 +-------------------------------------+-----+---+---+---+---+---+---+ 1324 | Enriching signaling messages | | | | | | | | 1325 | Signal features | 5.1 | | | | | x | | 1326 +-------------------------------------+-----+---+---+---+---+---+---+ 1328 6. Security Considerations 1330 This document has no requirement for a change to the security models 1331 within GMPLS and associated protocols. That is the OSPF-TE, RSVP-TE, 1332 and PCEP security models could be operated unchanged. 1334 However satisfying the requirements for RWA using the existing 1335 protocols may significantly affect the loading of those protocols. 1336 This makes the operation of the network more vulnerable to denial of 1337 service attacks. Therefore additional care maybe required to ensure 1338 that the protocols are secure in the WSON environment. 1340 Furthermore the additional information distributed in order to 1341 address the RWA problem represents a disclosure of network 1342 capabilities that an operator may wish to keep private. Consideration 1343 should be given to securing this information. 1345 7. IANA Considerations 1347 This document makes no request for IANA actions. 1349 8. Acknowledgments 1351 The authors would like to thank Adrian Farrel for many helpful 1352 comments that greatly improved the contents of this draft. 1354 This document was prepared using 2-Word-v2.0.template.dot. 1356 9. References 1358 9.1. Normative References 1360 [RFC3471] Berger, L., "Generalized Multi-Protocol Label Switching 1361 (GMPLS) Signaling Functional Description", RFC 3471, 1362 January 2003. 1364 [RFC3630] Katz, D., Kompella, K., and D. Yeung, "Traffic Engineering 1365 (TE) Extensions to OSPF Version 2", RFC 3630, September 1366 2003. 1368 [RFC3945] Mannie, E., "Generalized Multi-Protocol Label Switching 1369 (GMPLS) Architecture", RFC 3945, October 2004. 1371 [RFC4201] Kompella, K., Rekhter, Y., and L. Berger, "Link Bundling in 1372 MPLS Traffic Engineering (TE)", RFC 4201, October 2005. 1374 [RFC4202] Kompella, K. and Y. Rekhter, "Routing Extensions in Support 1375 of Generalized Multi-Protocol Label Switching (GMPLS)", RFC 1376 4202, October 2005. 1378 [RFC4328] Papadimitriou, D., "Generalized Multi-Protocol Label 1379 Switching (GMPLS) Signaling Extensions for G.709 Optical 1380 Transport Networks Control", RFC 4328, January 2006. 1382 [G.694.1] ITU-T Recommendation G.694.1, "Spectral grids for WDM 1383 applications: DWDM frequency grid", June, 2002. 1385 [RFC5088] J.L. Le Roux, J.P. Vasseur, Yuichi Ikejiri, and Raymond 1386 Zhang, "OSPF protocol extensions for Path Computation 1387 Element (PCE) Discovery", January 2008. 1389 [PCE-GCO] Y. Lee, J.L. Le Roux, D. King, and E. Oki, "Path 1390 Computation Element Communication Protocol (PCECP) 1391 Requirements and Protocol Extensions In Support of Global 1392 Concurrent Optimization", work in progress, draft-ietf-pce- 1393 global-concurrent-optimization-05.txt, November 2007. 1395 [PCEP] J.P. Vasseur and J.L. Le Roux (Editors), "Path Computation 1396 Element (PCE) Communication Protocol (PCEP)", work in 1397 progress, draft-ietf-pce-pcep-16.txt, February 2008. 1399 [PCE-OF] J.L. Le Roux, J.P. Vasseur, and Y. Lee, "Encoding of 1400 Objective Functions in Path Computation Element (PCE) 1401 communication and discovery protocols", work in progress, 1402 draft-ietf-pce-of-05.txt, February 2008. 1404 [WSON-Encode] G. Bernstein, Y. Lee, D. Li, and W. Imajuku, "Routing 1405 and Wavelength Assignment Information Encoding for 1406 Wavelength Switched Optical Networks", draft-bernstein- 1407 ccamp-wson-encode-00.txt, July 2008. 1409 [WSON-Info] G. Bernstein, Y. Lee, D. Li, W. Imajuku," Routing and 1410 Wavelength Assignment Information for Wavelength Switched 1411 Optical Networks", draft-bernstein-ccamp-wson-info-03.txt, 1412 July, 2008. 1414 9.2. Informative References 1416 [HZang00] H. Zang, J. Jue and B. Mukherjeee, "A review of routing and 1417 wavelength assignment approaches for wavelength-routed 1418 optical WDM networks", Optical Networks Magazine, January 1419 2000. 1421 [Coldren04] Larry A. Coldren, G. A. Fish, Y. Akulova, J. S. 1422 Barton, L. Johansson and C. W. Coldren, "Tunable 1423 Seiconductor Lasers: A Tutorial", Journal of Lightwave 1424 Technology, vol. 22, no. 1, pp. 193-202, January 2004. 1426 [Chu03] Xiaowen Chu, Bo Li and Chlamtac I, "Wavelength converter 1427 placement under different RWA algorithms in wavelength- 1428 routed all-optical networks", IEEE Transactions on 1429 Communications, vol. 51, no. 4, pp. 607-617, April 2003. 1431 [Buus06] Jens Buus EJM, "Tunable Lasers in Optical Networks", 1432 Journal of Lightware Technology, vol. 24, no. 1, pp. 5-11, 1433 January 2006. 1435 [Basch06] E. Bert Bash, Roman Egorov, Steven Gringeri and Stuart 1436 Elby, "Architectural Tradeoffs for Reconfigurable Dense 1437 Wavelength-Division Multiplexing Systems", IEEE Journal of 1438 Selected Topics in Quantum Electronics, vol. 12, no. 4, pp. 1439 615-626, July/August 2006. 1441 [Otani] T. Otani, H. Guo, K. Miyazaki, D. Caviglia, "Generalized 1442 Labels of Lambda-Switching Capable Label Switching Routers 1443 (LSR)", work in progress: draft-otani-ccamp-gmpls-lambda- 1444 labels-02.txt, November 2007. 1446 [Winzer06] Peter J. Winzer and Rene-Jean Essiambre, "Advanced 1447 Optical Modulation Formats", Proceedings of the IEEE, vol. 1448 94, no. 5, pp. 952-985, May 2006. 1450 [G.652] ITU-T Recommendation G.652, Characteristics of a single-mode 1451 optical fibre and cable, June 2005. 1453 [G.653] ITU-T Recommendation G.653, Characteristics of a dispersion- 1454 shifted single-mode optical fibre and cable, December 2006. 1456 [G.654] ITU-T Recommendation G.654, Characteristics of a cut-off 1457 shifted single-mode optical fibre and cable, December 2006. 1459 [G.655] ITU-T Recommendation G.655, Characteristics of a non-zero 1460 dispersion-shifted single-mode optical fibre and cable, 1461 March 2006. 1463 [G.656] ITU-T Recommendation G.656, Characteristics of a fibre and 1464 cable with non-zero dispersion for wideband optical 1465 transport, December 2006. 1467 [G.671] ITU-T Recommendation G.671, Transmission characteristics of 1468 optical components and subsystems, January 2005. 1470 [G.872] ITU-T Recommendation G.872, Architecture of optical 1471 transport networks, November 2001. 1473 [G.959.1] ITU-T Recommendation G.959.1, Optical Transport Network 1474 Physical Layer Interfaces, March 2006. 1476 [G.694.1] ITU-T Recommendation G.694.1, Spectral grids for WDM 1477 applications: DWDM frequency grid, June 2002. 1479 [G.694.2] ITU-T Recommendation G.694.2, Spectral grids for WDM 1480 applications: CWDM wavelength grid, December 2003. 1482 [G.Sup39] ITU-T Series G Supplement 39, Optical system design and 1483 engineering considerations, February 2006. 1485 [G.Sup43] ITU-T Series G Supplement 43, Transport of IEEE 10G base-R 1486 in optical transport networks (OTN), November 2006. 1488 [Imajuku] W. Imajuku, Y. Sone, I. Nishioka, S. Seno, "Routing 1489 Extensions to Support Network Elements with Switching 1490 Constraint", work in progress: draft-imajuku-ccamp-rtg- 1491 switching-constraint-02.txt, July 2007. 1493 [Ozdaglar03] Asuman E. Ozdaglar and Dimitri P. Bertsekas, ''Routing 1494 and wavelength assignment in optical networks,'' IEEE/ACM 1495 Transactions on Networking, vol. 11, 2003, pp. 259 -272. 1497 [RFC4054] Strand, J. and A. Chiu, "Impairments and Other Constraints 1498 on Optical Layer Routing", RFC 4054, May 2005. 1500 [RFC4606] Mannie, E. and D. Papadimitriou, "Generalized Multi- 1501 Protocol Label Switching (GMPLS) Extensions for Synchronous 1502 Optical Network (SONET) and Synchronous Digital Hierarchy 1503 (SDH) Control", RFC 4606, August 2006. 1505 10. Contributors 1507 Snigdho Bardalai 1508 Fujitsu 1509 Email: Snigdho.Bardalai@us.fujitsu.com 1511 Diego Caviglia 1512 Ericsson 1513 Via A. Negrone 1/A 16153 1514 Genoa Italy 1516 Phone: +39 010 600 3736 1517 Email: diego.caviglia@(marconi.com, ericsson.com) 1519 Daniel King 1520 Aria Networks 1521 Email: daniel.king@aria-networks.com 1523 Itaru Nishioka 1524 NEC Corp. 1525 1753 Simonumabe, Nakahara-ku, Kawasaki, Kanagawa 211-8666 1526 Japan 1527 Phone: +81 44 396 3287 1528 Email: i-nishioka@cb.jp.nec.com 1530 Lyndon Ong 1531 Ciena 1532 Email: Lyong@Ciena.com 1534 Pierre Peloso 1535 Alcatel-Lucent 1536 Route de Villejust - - 91620 Nozay - France 1537 Email: pierre.peloso@alcatel-lucent.fr 1539 Jonathan Sadler 1540 Tellabs 1541 Email: Jonathan.Sadler@tellabs.com 1543 Author's Addresses 1545 Greg M. Bernstein (ed.) 1546 Grotto Networking 1547 Fremont California, USA 1549 Phone: (510) 573-2237 1550 Email: gregb@grotto-networking.com 1551 Young Lee (ed.) 1552 Huawei Technologies 1553 1700 Alma Drive, Suite 100 1554 Plano, TX 75075 1555 USA 1557 Phone: (972) 509-5599 (x2240) 1558 Email: ylee@huawei.com 1560 Wataru Imajuku 1561 NTT Network Innovation Labs 1562 1-1 Hikari-no-oka, Yokosuka, Kanagawa 1563 Japan 1565 Phone: +81-(46) 859-4315 1566 Email: imajuku.wataru@lab.ntt.co.jp 1568 Intellectual Property Statement 1570 The IETF takes no position regarding the validity or scope of any 1571 Intellectual Property Rights or other rights that might be claimed to 1572 pertain to the implementation or use of the technology described in 1573 this document or the extent to which any license under such rights 1574 might or might not be available; nor does it represent that it has 1575 made any independent effort to identify any such rights. Information 1576 on the procedures with respect to rights in RFC documents can be 1577 found in BCP 78 and BCP 79. 1579 Copies of IPR disclosures made to the IETF Secretariat and any 1580 assurances of licenses to be made available, or the result of an 1581 attempt made to obtain a general license or permission for the use of 1582 such proprietary rights by implementers or users of this 1583 specification can be obtained from the IETF on-line IPR repository at 1584 http://www.ietf.org/ipr. 1586 The IETF invites any interested party to bring to its attention any 1587 copyrights, patents or patent applications, or other proprietary 1588 rights that may cover technology that may be required to implement 1589 this standard. Please address the information to the IETF at 1590 ietf-ipr@ietf.org. 1592 Disclaimer of Validity 1594 This document and the information contained herein are provided on an 1595 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 1596 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 1597 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 1598 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 1599 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 1600 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 1602 Copyright Statement 1604 Copyright (C) The IETF Trust (2008). 1606 This document is subject to the rights, licenses and restrictions 1607 contained in BCP 78, and except as set forth therein, the authors 1608 retain all their rights. 1610 Acknowledgment 1612 Funding for the RFC Editor function is currently provided by the 1613 Internet Society.