idnits 2.17.1 draft-ietf-ccamp-rwa-wson-framework-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? -- It seems you're using the 'non-IETF stream' Licence Notice instead Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (March 4, 2009) is 5532 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== Unused Reference: 'G.872' is defined on line 1907, but no explicit reference was found in the text == Outdated reference: A later version (-10) exists of draft-ietf-pce-global-concurrent-optimization-08 Summary: 1 error (**), 0 flaws (~~), 3 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 Network Working Group Y. Lee (ed.) 2 Internet Draft Huawei 3 Intended status: Informational G. Bernstein (ed.) 4 Expires: September 2009 Grotto Networking 5 Wataru Imajuku 6 NTT 8 March 4, 2009 10 Framework for GMPLS and PCE Control of Wavelength Switched Optical 11 Networks (WSON) 12 draft-ietf-ccamp-rwa-wson-framework-02.txt 14 Status of this Memo 16 This Internet-Draft is submitted to IETF in full conformance with the 17 provisions of BCP 78 and BCP 79. 19 Internet-Drafts are working documents of the Internet Engineering 20 Task Force (IETF), its areas, and its working groups. Note that 21 other groups may also distribute working documents as Internet- 22 Drafts. 24 Internet-Drafts are draft documents valid for a maximum of six months 25 and may be updated, replaced, or obsoleted by other documents at any 26 time. It is inappropriate to use Internet-Drafts as reference 27 material or to cite them other than as "work in progress." 29 The list of current Internet-Drafts can be accessed at 30 http://www.ietf.org/ietf/1id-abstracts.txt 32 The list of Internet-Draft Shadow Directories can be accessed at 33 http://www.ietf.org/shadow.html 35 This Internet-Draft will expire on September 4, 2009. 37 Copyright Notice 39 Copyright (c) 2009 IETF Trust and the persons identified as the 40 document authors. All rights reserved. 42 This document is subject to BCP 78 and the IETF Trust's Legal 43 Provisions Relating to IETF Documents 44 (http://trustee.ietf.org/license-info) in effect on the date of 45 publication of this document. Please review these documents 46 carefully, as they describe your rights and restrictions with respect 47 to this document. 49 Abstract 51 This memo provides a framework for applying Generalized Multi- 52 Protocol Label Switching (GMPLS) and the Path Computation Element 53 (PCE) architecture to the control of wavelength switched optical 54 networks (WSON). In particular we provide control plane models for 55 key wavelength switched optical network subsystems and processes. The 56 subsystems include wavelength division multiplexed links, tunable 57 laser transmitters, reconfigurable optical add/drop multiplexers 58 (ROADM) and wavelength converters. 60 Lightpath provisioning, in general, requires the routing and 61 wavelength assignment (RWA) process. This process is reviewed and the 62 information requirements, both static and dynamic for this process 63 are presented, along with alternative implementation scenarios that 64 could be realized via GMPLS/PCE and/or extended GMPLS/PCE protocols. 65 This memo does NOT address optical impairments in any depth and 66 focuses on topological elements and path selection constraints that 67 are common across different WSON environments. It is expected that a 68 variety of different techniques will be applied to optical 69 impairments depending on the type of WSON, such as access, metro or 70 long haul. 72 Table of Contents 74 1. Introduction...................................................4 75 1.1. Revision History..........................................5 76 1.1.1. Changes from 00......................................5 77 1.1.2. Changes from 01......................................5 78 2. Terminology....................................................5 79 3. Wavelength Switched Optical Networks...........................6 80 3.1. WDM and CWDM Links........................................6 81 3.2. Optical Transmitters......................................8 82 3.2.1. Lasers...............................................8 83 3.2.2. Spectral Characteristics & Modulation Type...........9 84 3.2.3. Signal Rates and Error Correction...................10 85 3.3. ROADMs, OXCs, Splitters, Combiners and FOADMs............10 86 3.3.1. Reconfigurable Add/Drop Multiplexers and OXCs.......11 87 3.3.2. Splitters...........................................13 88 3.3.3. Combiners...........................................13 89 3.3.4. Fixed Optical Add/Drop Multiplexers.................13 90 3.4. Wavelength Converters....................................14 91 3.4.1. Wavelength Converter Pool Modeling..................16 92 4. Routing and Wavelength Assignment and the Control Plane.......20 93 4.1. Architectural Approaches to RWA..........................21 94 4.1.1. Combined RWA (R&WA).................................22 95 4.1.2. Separated R and WA (R+WA)...........................22 96 4.1.3. Routing and Distributed WA (R+DWA)..................23 97 4.2. Conveying information needed by RWA......................23 98 4.3. Lightpath Temporal Characteristics.......................24 99 5. Modeling Examples and Control Plane Use Cases.................25 100 5.1. Network Modeling for GMPLS/PCE Control...................25 101 5.1.1. Describing the WSON nodes...........................26 102 5.1.2. Describing the links................................28 103 5.2. RWA Path Computation and Establishment...................29 104 5.3. Resource Optimization....................................30 105 5.4. Support for Rerouting....................................31 106 6. GMPLS & PCE Implications......................................31 107 6.1. Implications for GMPLS signaling.........................31 108 6.1.1. Identifying Wavelengths and Signals.................32 109 6.1.2. Combined RWA/Separate Routing WA support............32 110 6.1.3. Distributed Wavelength Assignment: Unidirectional, No 111 Converters.................................................32 112 6.1.4. Distributed Wavelength Assignment: Unidirectional, 113 Limited Converters.........................................33 114 6.1.5. Distributed Wavelength Assignment: Bidirectional, No 115 Converters.................................................34 116 6.2. Implications for GMPLS Routing...........................34 117 6.2.1. Need for Wavelength-Specific Maximum Bandwidth 118 Information................................................35 119 6.2.2. Need for Wavelength-Specific Availability Information35 120 6.2.3. Relationship to Link Bundling and Layering..........36 121 6.2.4. WSON Routing Information Summary....................36 122 6.3. Optical Path Computation and Implications for PCE........37 123 6.3.1. Lightpath Constraints and Characteristics...........37 124 6.3.2. Computation Architecture Implications...............38 125 6.3.3. Discovery of RWA Capable PCEs.......................38 126 6.4. Scaling Implications.....................................39 127 6.4.1. Routing.............................................39 128 6.4.2. Signaling...........................................39 129 6.4.3. Path computation....................................39 130 6.5. Summary of Impacts by RWA Architecture...................40 131 7. Security Considerations.......................................41 132 8. IANA Considerations...........................................41 133 9. Acknowledgments...............................................41 134 10. References...................................................42 135 10.1. Normative References....................................42 136 10.2. Informative References..................................43 137 11. Contributors.................................................46 138 Author's Addresses...............................................46 139 Intellectual Property Statement..................................47 140 Disclaimer of Validity...........................................48 142 1. Introduction 144 From its beginning Generalized Multi-Protocol Label Switching (GMPLS) 145 was intended to control wavelength switched optical networks (WSON) 146 with the GMPLS architecture document [RFC3945] explicitly mentioning 147 both wavelength and waveband switching and equating wavelengths 148 (lambdas) with GMPLS labels. In addition a discussion of optical 149 impairments and other constraints on optical routing can be found in 150 [RFC4054]. However, optical technologies have advanced in ways that 151 make them significantly different from other circuit switched 152 technologies such as Time Division Multiplexing (TDM). Service 153 providers have already deployed many of these new optical 154 technologies such as ROADMs and tunable lasers and desire the same 155 automation and restoration capabilities that GMPLS has provided to 156 TDM and packet switched networks. Another important application of an 157 automated control plane such as GMPLS is the possibility to improve, 158 via recovery schemes, the availability of the network. One of the 159 key points of GMPLS based recovery schemes is the capability to 160 survive multiple failures while legacy protection mechanism such as 161 1+1 path protection can survive from a single failure. Moreover this 162 improved availability can be obtained using less network resources. 164 This document will focus on the unique properties of links, switches 165 and path selection constraints that occur in WSONs. Different WSONs 166 such as access, metro and long haul may apply different techniques 167 for dealing with optical impairments hence this document will NOT 168 address optical impairments in any depth, but instead focus on 169 properties that are common across a variety of WSONs. 171 This memo provides a framework for applying GMPLS and the Path 172 Computation Element (PCE) architecture to the control of WSONs. In 173 particular we provide control plane models for key wavelength 174 switched optical network subsystems and processes. The subsystems 175 include wavelength division multiplexed links, tunable laser 176 transmitters, reconfigurable optical add/drop multiplexers (ROADM) 177 and wavelength converters. 179 Lightpath provisioning, in general, requires the routing and 180 wavelength assignment (RWA) process. This process is reviewed and the 181 information requirements, both static and dynamic for this process 182 are presented, along with alternative implementation architectures 183 that could be realized via various combinations of extended GMPLS and 184 PCE protocols. 186 1.1. Revision History 188 1.1.1. Changes from 00 190 o Added new first level section on modeling examples and control 191 plane use cases. 193 o Added new third level section on wavelength converter pool 194 modeling 196 o Editorial clean up of English and updated references. 198 1.1.2. Changes from 01 200 Fixed error in wavelength converter pool example. 202 2. Terminology 204 CWDM: Coarse Wavelength Division Multiplexing. 206 DWDM: Dense Wavelength Division Multiplexing. 208 FOADM: Fixed Optical Add/Drop Multiplexer. 210 OXC: Optical cross connect. A symmetric optical switching element in 211 which a signal on any ingress port can reach any egress port. 213 ROADM: Reconfigurable Optical Add/Drop Multiplexer. An asymmetric 214 wavelength selective switching element featuring ingress and egress 215 line side ports as well as add/drop side ports. 217 RWA: Routing and Wavelength Assignment. 219 Wavelength Conversion/Converters: The process of converting an 220 information bearing optical signal centered at a given wavelength to 221 one with "equivalent" content centered at a different wavelength. 222 Wavelength conversion can be implemented via an optical-electronic- 223 optical (OEO) process or via a strictly optical process. 225 WDM: Wavelength Division Multiplexing. 227 Wavelength Switched Optical Networks (WSON): WDM based optical 228 networks in which switching is performed selectively based on the 229 center wavelength of an optical signal. 231 3. Wavelength Switched Optical Networks 233 WSONs come in a variety of shapes and sizes from continent spanning 234 long haul networks, to metropolitan networks, to residential access 235 networks. In all these cases we are concerned with those properties 236 that constrain the choice of wavelengths that can be used, i.e., 237 restrict the wavelength label set, impact the path selection process, 238 and limit the topological connectivity. In the following we examine 239 and model some major subsystems of a WSON with an emphasis on those 240 aspects that are of relevance to the control plane. In particular we 241 look at WDM links, Optical Transmitters, ROADMs, and Wavelength 242 Converters. 244 3.1. WDM and CWDM Links 246 WDM and CWDM links run over optical fibers, and optical fibers come 247 in a wide range of types that tend to be optimized for various 248 applications from access networks, metro, long haul, and submarine 249 links to name a few. ITU-T and IEC standards exist for various types 250 of fibers. For the purposes here we are concerned only with single 251 mode fibers (SMF). The following SMF fiber types are typically 252 encountered in optical networks: 254 ITU-T Standard | Common Name 255 ------------------------------------------------------------ 256 G.652 [G.652] | Standard SMF | 257 G.653 [G.653] | Dispersion shifted SMF | 258 G.654 [G.654] | Cut-off shifted SMF | 259 G.655 [G.655] | Non-zero dispersion shifted SMF | 260 G.656 [G.656] | Wideband non-zero dispersion shifted SMF | 261 ------------------------------------------------------------ 262 These fiber types are differentiated by their optical impairment 263 characteristics such as attenuation, chromatic dispersion, 264 polarization mode dispersion, four wave mixing, etc. Since these 265 effects can be dependent upon wavelength, channel spacing and input 266 power level, the net effect for our modeling purposes here is to 267 restrict the range of wavelengths that can be used. 269 Typically WDM links operate in one or more of the approximately 270 defined optical bands [G.Sup39]: 272 Band Range (nm) Common Name Raw Bandwidth (THz) 273 O-band 1260-1360 Original 17.5 274 E-band 1360-1460 Extended 15.1 275 S-band 1460-1530 Short 9.4 276 C-band 1530-1565 Conventional 4.4 277 L-band 1565-1625 Long 7.1 278 U-band 1625-1675 Ultra-long 5.5 280 Not all of a band may be usable, for example in many fibers that 281 support E-band there is significant attenuation due to a water 282 absorption peak at 1383nm. Hence we can have a discontinuous 283 acceptable wavelength range for a particular link. Also some systems 284 will utilize more than one band. This is particularly true for coarse 285 WDM (CWDM) systems. 287 [Editor's note: the previous text is primarily tutorial in nature and 288 maybe deleted or moved to an appendix in a future draft] 290 Current technology breaks up the bandwidth capacity of fibers into 291 distinct channels based on either wavelength or frequency. There are 292 two standards covering wavelengths and channel spacing. ITU-T 293 recommendation [G.694.1] describes a DWDM grid defined in terms of 294 frequency grids of 12.5GHz, 25GHz, 50GHz, 100GHz, and other multiples 295 of 100GHz around a 193.1THz center frequency. At the narrowest 296 channel spacing this provides less than 4800 channels across the O 297 through U bands. ITU-T recommendation [G.694.2] describes a CWDM grid 298 defined in terms of wavelength increments of 20nm running from 1271nm 299 to 1611nm for 18 or so channels. The number of channels is 300 significantly smaller than the 32 bit GMPLS label space allocated to 301 lambda switching. A label representation for these ITU-T grids is 302 given in [Otani] and allows a common vocabulary to be used in 303 signaling lightpaths. Further, these ITU-T grid based labels can and 304 also be used to describe WDM links, ROADM ports, and wavelength 305 converters for the purposes path selection. 307 With a tremendous existing base of fiber many WDM links are designed 308 to take advantage of particular fiber characteristics or to try to 309 avoid undesirable properties. For example dispersion shifted SMF 310 [G.653] was originally designed for good long distance performance in 311 single channel systems, however putting WDM over this type of fiber 312 requires much system engineering and a fairly limited range of 313 wavelengths. Hence for our basic, impairment unaware, modeling of a 314 WDM link we will need the following information: 316 o Wavelength range(s): Given a mapping between labels and the ITU-T 317 grids each range could be expressed in terms of a doublet 318 (lambda1, lambda2) or (freq1, freq1) where the lambdas or 319 frequencies can be represented by 32 bit integers. 321 o Channel spacing: currently there are about five channel spacings 322 used in DWDM systems 12.5GHz to 200GHz and one defined CWDM 323 spacing. 325 For a particular link this information is relatively static, i.e., 326 changes to these properties generally require hardware upgrades. Such 327 information could be used locally during wavelength assignment via 328 signaling, similar to label restrictions in MPLS or used by a PCE in 329 solving the combined routing and wavelength assignment problem. 331 3.2. Optical Transmitters 333 3.2.1. Lasers 335 WDM optical systems make use of laser transmitters utilizing 336 different wavelengths (frequencies). Some laser transmitters were and 337 are manufactured for a specific wavelength of operation, that is, the 338 manufactured frequency cannot be changed. First introduced to reduce 339 inventory costs, tunable optical laser transmitters are becoming 340 widely deployed in some systems [Coldren04], [Buus06]. This allows 341 flexibility in the wavelength used for optical transmission and aids 342 in the control of path selection. 344 Fundamental modeling parameters from the control plane perspective 345 optical transmitters are: 347 o Tunable: Is this transmitter tunable or fixed. 349 o Tuning range: This is the frequency or wavelength range over which 350 the laser can be tuned. With the fixed mapping of labels to 351 lambdas of [Otani] this can be expressed as a doublet (lambda1, 352 lambda2) or (freq1, freq2) where lambda1 and lambda2 or freq1 and 353 freq2 are the labels representing the lower and upper bounds in 354 wavelength or frequency. 356 o Tuning time: Tuning times highly depend on the technology used. 357 Thermal drift based tuning may take seconds to stabilize, whilst 358 electronic tuning might provide sub-ms tuning times. Depending on 359 the application this might be critical. For example, thermal drift 360 might not be applicable for fast protection applications. 362 o Spectral Characteristics and stability: The spectral shape of the 363 laser's emissions and its frequency stability put limits on 364 various properties of the overall WDM system. One relatively easy 365 to characterize constraint is the finest channel spacing on which 366 the transmitter can be used. 368 Note that ITU-T recommendations specify many other aspects of a 369 laser's such as spectral characteristics and stability. Many of these 370 parameters are used in the design of WDM subsystems consisting of 371 transmitters, WDM links and receivers however they do not furnish 372 additional information that will influence label switched path (LSP) 373 provisioning in a properly designed system. 375 Also note that lasers transmitters as a component can degrade and 376 fail over time. This presents the possibility of the failure of a LSP 377 (lightpath) without either a node or link failure. Hence, additional 378 mechanisms may be necessary to detect and differentiate this failure 379 from the others, e.g., one doesn't not want to initiate mesh 380 restoration if the source transmitter has failed, since the laser 381 transmitter will still be failed on the alternate optical path. 383 3.2.2. Spectral Characteristics & Modulation Type 385 Contrary to some marketing claims optical systems are not truly 386 "transparent" to the content of the signals that they carry. Each 387 lightpath will have spectral characteristics based on its content, 388 and the spacing of wavelengths in a WDM link will ultimately put 389 constraints on that spectrum. 391 For analog signals such as used in closed access television (CATV) or 392 "radio over fiber" links spectral characteristics are given in terms 393 of various bandwidth measures. However digital signals consist of our 394 main focus here and in the ITU-T G series optical specifications. In 395 this case the spectral characteristics can be more accurately 396 inferred from the modulation format and the bit rate. 398 Although Non-Return to Zero (NRZ) is currently the dominant form of 399 optical modulation, new modulation formats are being researched 400 [Winzer06] and deployed. With a choice in modulation formats we no 401 longer have a one to one relationship between digital bandwidth in 402 bytes or bits per second and the amount of optical spectrum (optical 403 bandwidth) consumed. To simplify the specification of optical signals 404 the ITU-T, in recommendation G.959.1, combined a rate bound and 405 modulation format designator [G.959.1]. For example, two of the 406 signal classes defined in [G.959.1] are: 408 Optical tributary signal class NRZ 1.25G: 410 "Applies to continuous digital signals with non-return to zero line 411 coding, from nominally 622 Mbit/s to nominally 1.25 Gbit/s. Optical 412 tributary signal class NRZ 1.25G includes a signal with STM-4 bit 413 rate according to ITU-T Rec. G.707/Y.1322." Note that Gigabit 414 Ethernet falls into this signaling class as well. 416 Optical tributary signal class RZ 40G: 418 "Applies to continuous digital signals with return to zero line 419 coding, from nominally 9.9 Gbit/s to nominally 43.02 Gbit/s. 420 Optical tributary signal class RZ 40G includes a signal with STM- 421 256 bit rate according to ITU-T Rec. G.707/Y.1322 and OTU3 bit rate 422 according to ITU-T Rec. G.709/Y.1331." 424 From a modeling perspective we have: 426 o Analog signals: bandwidth parameters, e.g., 3dB parameters and 427 similar. 429 o Digital signals: there are predefined modulation bit rate classes 430 that we can encode. 432 This information can be important in constraining route selection, 433 for example some signals may not be compatible with some links or 434 wavelength converters. In addition it lets the endpoints understand 435 if it can process the signal. 437 3.2.3. Signal Rates and Error Correction 439 Although, the spectral characteristics of a signal determine its 440 basic compatibility with a WDM system, more information is generally 441 needed for various processing activities such as regeneration and 442 reception. Many digital signals such as Ethernet, G.709, and SDH have 443 well defined encoding which includes forward error correction (FEC). 444 However many subsystem vendors offer additional FEC options for a 445 given signal type. The use of different FECs can lead to different 446 overall signal rates. If the FEC and rate used is not compatible 447 between the sender and receiver the signal can not be correctly 448 processed. Note that the rates of "standard" signals may be extended 449 to accommodate different payloads. For example there are 450 transmitters capable of directly mapping 10GE LAN-PHY traffic into 451 G.709 ODU2 frame with slightly higher clock rate [G.Sup43]. 453 3.3. ROADMs, OXCs, Splitters, Combiners and FOADMs 455 Definitions of various optical devices and their parameters can be 456 found in [G.671], we only look at a subset of these and their non- 457 impairment related properties. 459 3.3.1. Reconfigurable Add/Drop Multiplexers and OXCs 461 Reconfigurable add/drop optical multiplexers (ROADM) have matured and 462 are available in different forms and technologies [Basch06]. This is 463 a key technology that allows wavelength based optical switching. A 464 classic degree-2 ROADM is shown in Figure 1. 466 Line side ingress +---------------------+ Line side egress 467 --->| |---> 468 | | 469 | ROADM | 470 | | 471 | | 472 +---------------------+ 473 | | | | o o o o 474 | | | | | | | | 475 O O O O | | | | 476 Tributary Side: Drop (egress) Add (ingress) 478 Figure 1 Degree-2 ROADM 480 The key feature across all ROADM types is their highly asymmetric 481 switching capability. In the ROADM of Figure 1, the "add" ingress 482 ports can only egress on the line side egress port and not on any of 483 the "drop" egress ports. The degree of a ROADM or switch is given by 484 the number of line side ports (ingress and egress) and does not 485 include the number of "add" or "drop" ports. Sometimes the "add" 486 "drop" ports are also called tributary ports. As the degree of the 487 ROADM increases beyond two it can have properties of both a switch 488 (OXC) and a multiplexer and hence we must know the switched 489 connectivity offered by such a network element to effectively utilize 490 it. A straight forward way to do this is via a "switched 491 connectivity" matrix A where Amn = 0 or 1, depending upon whether a 492 wavelength on ingress port m can be connected to egress port n 493 [Imajuku]. For the ROADM of Figure 1 the switched connectivity matrix 494 can be expressed as 495 Ingress Egress Port 496 Port #1 #2 #3 #4 #5 497 -------------- 498 #1: 1 1 1 1 1 499 #2 1 0 0 0 0 500 A = #3 1 0 0 0 0 501 #4 1 0 0 0 0 502 #5 1 0 0 0 0 504 Where ingress ports 2-5 are add ports, egress ports 2-5 are drop 505 ports and ingress port #1 and egress port #1 are the line side (WDM) 506 ports. 508 For ROADMs this matrix will be very sparse, and for OXCs the 509 complement of the matrix will be very sparse, compact encodings and 510 usage including high degree ROADMs/OXCs are given in [WSON-Encode]. 512 Additional constraints may also apply to the various ports in a 513 ROADM/OXC. In the literature of optical switches and ROADMs the 514 following restrictions/terms are used: 516 Colored port: An ingress or more typically an egress (drop) port 517 restricted to a single channel of fixed wavelength. 519 Colorless port: An ingress or more typically an egress (drop) port 520 restricted to a single channel of arbitrary wavelength. 522 In general a port on a ROADM could have any of the following 523 wavelength restrictions: 525 o Multiple wavelengths, full range port 527 o Single wavelength, full range port 529 o Single wavelength, fixed lambda port 531 o Multiple wavelengths, reduced range port (for example wave band 532 switching) 534 To model these restrictions we need two pieces of information for 535 each port: (a) number of wavelengths, (b) wavelength range and 536 spacing. Note that this information is relatively static. More 537 complicated wavelength constraints are modeled in [WSON-Info]. 539 3.3.2. Splitters 541 An optical splitter consists of a single ingress port and two or more 542 egress ports. The ingress optical signaled is essentially copied 543 (with loss) to all egress ports. 545 Using the modeling notions of section 3.3.1. the ingress and egress 546 ports of a splitter would have the same wavelength restrictions. In 547 addition we can describe a splitter by a connectivity matrix Amn as 548 follows: 550 Ingress Egress Port 551 Port #1 #2 #3 ... #N 552 ----------------- 553 A = #1 1 1 1 ... 1 555 The difference from a simple ROADM is that this is not a switched 556 connectivity matrix but the fixed connectivity matrix of the device. 558 3.3.3. Combiners 560 A optical combiner is somewhat the dual of a splitter in that it has 561 a single multi-wavelength egress port and multiple ingress ports. 562 The contents of all the ingress ports are copied and combined to the 563 single egress port. The various ports may have different wavelength 564 restrictions. It is generally the responsibility of those using the 565 combiner to assure that wavelength collision does not occur on the 566 egress port. The fixed connectivity matrix Amn for a combiner would 567 look like: 569 Ingress Egress Port 570 Port #1 571 --- 572 #1: 1 573 #2 1 574 A = #3 1 575 ... 1 576 #N 1 578 3.3.4. Fixed Optical Add/Drop Multiplexers 580 A fixed optical add/drop multiplexer can alter the course of an 581 ingress wavelength in a preset way. In particular a particular 582 wavelength (or waveband) from a line side ingress port would be 583 dropped to a particular "tributary" egress port. Depending on the 584 device's fixed configuration that same wavelength may or may not be 585 "continued" to the line side egress port ("drop and continue" 586 operation). Further there may exist tributary ingress ports ("add" 587 ports) whose signals are combined with each other and "continued" 588 line side signals. 590 In general to represent the routing properties of an FOADM we need a 591 fixed connectivity matrix Amn as previously discussed and we need the 592 precise wavelength restrictions for all ingress and egress ports. 593 From the wavelength restrictions on the tributary egress ports (drop 594 ports) we can see what wavelengths have been dropped. From the 595 wavelength restrictions on the tributary ingress (add) ports we can 596 see which wavelengths have been added to the line side egress port. 597 Finally from the added wavelength information and the line side 598 egress wavelength restrictions we can infer which wavelengths have 599 been continued. 601 To summarize, the modeling methodology introduced in section 3.3.1. 602 consisting of a connectivity matrix and port wavelength restrictions 603 can be used to describe a large set of fixed optical devices such as 604 combiners, splitters and FOADMs. Hybrid devices consisting of both 605 switched and fixed parts are modeled in [WSON-Info]. 607 3.4. Wavelength Converters 609 Wavelength converters take an ingress optical signal at one 610 wavelength and emit an equivalent content optical signal at another 611 wavelength on egress. There are currently two approaches to building 612 wavelength converters. One approach is based on optical to electrical 613 to optical (OEO) conversion with tunable lasers on egress. This 614 approach can be dependent upon the signal rate and format, i.e., this 615 is basically an electrical regenerator combined with a tunable laser. 616 The other approach performs the wavelength conversion, optically via 617 non-linear optical effects, similar in spirit to the familiar 618 frequency mixing used in radio frequency systems, but significantly 619 harder to implement. Such processes/effects may place limits on the 620 range of achievable conversion. These may depend on the wavelength of 621 the input signal and the properties of the converter as opposed to 622 only the properties of the converter in the OEO case. Different WSON 623 system designs may choose to utilize this component to varying 624 degrees or not at all. 626 Current or envisioned contexts for wavelength converters are: 628 1. Wavelength conversion associated with OEO switches and tunable 629 laser transmitters. In this case there are plenty of converters to 630 go around since we can think of each tunable output laser 631 transmitter on an OEO switch as a potential wavelength converter. 633 2. Wavelength conversion associated with ROADMs/OXCs. In this case we 634 may have a limited amount of conversion available. Conversion could 635 be either all optical or via an OEO method. 637 3. Wavelength conversion associated with fixed devices such as FOADMs. 638 In this case we may have a limited amount of conversion. Also in 639 this case the conversion may be used as part of light path routing. 641 Based on the above contexts a tentative modeling approach for 642 wavelength converters could be as follows: 644 1. Wavelength converters can always be modeled as associated with 645 network elements. This includes fixed wavelength routing elements. 647 2. A network element may have full wavelength conversion capability, 648 i.e., any ingress port and wavelength, or a limited number of 649 wavelengths and ports. On a box with a limited number of 650 converters there also may exist restrictions on which ports can 651 reach the converters. Hence regardless of where the converters 652 actually are we can associate them with ingress ports. 654 3. Wavelength converters have range restrictions that are either 655 independent or dependent upon the ingress wavelength. [TBD: for 656 those that depend on ingress wavelength can we have a standard 657 formula? Also note that this type of converter introduces 658 additional optical impairments.] 660 4. Wavelength converters that are O-E-O based will have a restriction 661 based on the modulation format and transmission speed. 663 Note that since O-E-O wavelength converters also serve as 664 regenerators we can include regenerators in our model of wavelength 665 converters. O-E-O Regenerators come in three general types known as 666 1R, 2R, and 3R regenerators. 1R regenerators re-amplify the signal to 667 combat attenuation, 2R regenerators reshape as well as amplify the 668 signal, 3R regenerators amplify, reshape and retime the signal. As we 669 go from 1R to 3R regenerators the signal is ''cleaned up'' better but 670 at the same time the regeneration process becomes more dependent on 671 the signal characteristics such as format and rate. 673 In WSONs where wavelength converters are sparse we may actually see a 674 light path appear to loop or ''backtrack'' upon itself in order to 675 reach a wavelength converter prior to continuing on to its 676 destination. The lambda used on the "detour" out to the wavelength 677 converter would be different from that coming back from the "detour" 678 to the wavelength converter. 680 A model for an individual O-E-O wavelength converter would consist 681 of: 683 o Input lambda or frequency range 685 o Output lambda or frequency range 687 o Equivalent regeneration level (1R, 2R, 3R) 689 o Signal restrictions if a 2R or 3R regeneration: formats and rates 691 [FFS: Model for an all optical wavelength converter] 693 3.4.1. Wavelength Converter Pool Modeling 695 A WSON node may include multiple wavelength converters. These are 696 usually arranged into some type of pool to promote resource sharing. 697 There are a number of different approaches used in the design of 698 switches with converter pools. However, from the point of view of 699 path computation we need to know the following: 701 1. The nodes that support wavelength conversion. 703 2. The accessibility and availability of a wavelength converter to 704 convert from a given ingress wavelength on a particular ingress 705 port to a desired egress wavelength on a particular egress port. 707 3. Limitations on the types of signals that can be converted and the 708 conversions that can be performed. 710 To model point 2 above we can use a similar technique as used to 711 model ROADMs and optical switches, i.e., a matrices to indicate 712 possible connectivity along with wavelength constraints for 713 links/ports. Since wavelength converters are considered a scarce 714 resource we will also want our model to include as a minimum the 715 usage state of individual wavelength converters in the pool. Models 716 that incorporate more state to further reveal blocking conditions on 717 ingress or egress to particular converters are for further study. 719 We utilize a three stage model as shown schematically in Figure 2. In 720 this model we assume N ingress ports (fibers), P wavelength 721 converters, and M egress ports (fibers). Since not all ingress ports 722 can necessarily reach the converter pool, the model starts with a 723 wavelength pool ingress matrix WI(i,p) = {0,1} whether ingress port i 724 can reach potentially reach wavelength converter p. 726 Since not all wavelength can necessarily reach all the converters or 727 the converters may have limited input wavelength range we have a set 728 of ingress port constraints for each wavelength converter. Currently 729 we assume that a wavelength converter can only take a single 730 wavelength on input. We can model each wavelength converter ingress 731 port constraint via a wavelength set mechanism. 733 Next we have a state vector WC(j) = {0,1} dependent upon whether 734 wavelength converter j in the pool is in use. This is the only state 735 kept in the converter pool model. This state is not necessary for 736 modeling "fixed" transponder system, i.e., systems where there is no 737 sharing. In addition, this state information may be encoded in a 738 much more compact form depending on the overall connectivity 739 structure [WC-Pool]. 741 After that, we have a set of wavelength converter egress wavelength 742 constraints. These constraints indicate what wavelengths a particular 743 wavelength converter can generate or are restricted to generating due 744 to internal switch structure. 746 Finally, we have a wavelength pool egress matrix WE(p,k) = {0,1} 747 depending on whether the output from wavelength converter p can reach 748 egress port k. Examples of this method being used to model wavelength 749 converter pools for several switch architectures from the literature 750 are given in reference [WC-Pool]. 752 I1 +-------------+ +-------------+ E1 753 ----->| | +--------+ | |-----> 754 I2 | +------+ WC #1 +-------+ | E2 755 ----->| | +--------+ | |-----> 756 | Wavelength | | Wavelength | 757 | Converter | +--------+ | Converter | 758 | Pool +------+ WC #2 +-------+ Pool | 759 | | +--------+ | | 760 | Ingress | | Egress | 761 | Connection | . | Connection | 762 | Matrix | . | Matrix | 763 | | . | | 764 | | | | 765 IN | | +--------+ | | EM 766 ----->| +------+ WC #P +-------+ |-----> 767 | | +--------+ | | 768 +-------------+ ^ ^ +-------------+ 769 | | 770 | | 771 | | 772 | | 774 Ingress wavelength Egress wavelength 775 constraints for constraints for 776 each converter each converter 778 Figure 2 Schematic diagram of wavelength converter pool model. 780 Example: Shared Per Node 782 In Figure 3 below we show a simple optical switch in a four 783 wavelength DWDM system sharing wavelength converters in a general 784 "per node" fashion. 786 ___________ +------+ 787 | |--------------------------->| | 788 | |--------------------------->| C | 789 /| | |--------------------------->| o | E1 790 I1 /D+--->| |--------------------------->| m | 791 + e+--->| | | b |====> 792 ====>| M| | Optical | +-----------+ +----+ | i | 793 + u+--->| Switch | | WC Pool | |O S|-->| n | 794 \x+--->| | | +-----+ | |p w|-->| e | 795 \| | +----+->|WC #1|--+->|t i| | r | 796 | | | +-----+ | |i t| +------+ 797 | | | | |c c| +------+ 798 /| | | | +-----+ | |a h|-->| | 799 I2 /D+--->| +----+->|WC #2|--+->|l |-->| C | E2 800 + e+--->| | | +-----+ | | | | o | 801 ====>| M| | | +-----------+ +----+ | m |====> 802 + u+--->| | | b | 803 \x+--->| |--------------------------->| i | 804 \| | |--------------------------->| n | 805 | |--------------------------->| e | 806 |___________|--------------------------->| r | 807 +------+ 809 Figure 3 An optical switch featuring a shared per node wavelength 810 converter pool architecture. 812 In this case the ingress and egress pool matrices are simply: 814 +-----+ +-----+ 815 | 1 1 | | 1 1 | 816 WI =| |, WE =| | 817 | 1 1 | | 1 1 | 818 +-----+ +-----+ 820 Example: Shared Per Link 822 In Figure 4 we show a different wavelength pool architecture know as 823 "shared per fiber". In this case the ingress and egress pool matrices 824 are simply: 826 +-----+ +-----+ 827 | 1 1 | | 1 0 | 828 WI =| |, WE =| | 829 | 1 1 | | 0 1 | 830 +-----+ +-----+ 832 ___________ +------+ 833 | |--------------------------->| | 834 | |--------------------------->| C | 835 /| | |--------------------------->| o | E1 836 I1 /D+--->| |--------------------------->| m | 837 + e+--->| | | b |====> 838 ====>| M| | Optical | +-----------+ | i | 839 + u+--->| Switch | | WC Pool | | n | 840 \x+--->| | | +-----+ | | e | 841 \| | +----+->|WC #1|--+---------->| r | 842 | | | +-----+ | +------+ 843 | | | | +------+ 844 /| | | | +-----+ | | | 845 I2 /D+--->| +----+->|WC #2|--+---------->| C | E2 846 + e+--->| | | +-----+ | | o | 847 ====>| M| | | +-----------+ | m |====> 848 + u+--->| | | b | 849 \x+--->| |--------------------------->| i | 850 \| | |--------------------------->| n | 851 | |--------------------------->| e | 852 |___________|--------------------------->| r | 853 +------+ 854 Figure 4 An optical switch featuring a shared per fiber wavelength 855 converter pool architecture. 857 4. Routing and Wavelength Assignment and the Control Plane 859 In wavelength switched optical networks consisting of tunable lasers 860 and wavelength selective switches with wavelength converters on every 861 interface, path selection is similar to the MPLS and TDM circuit 862 switched cases in that the labels, in this case wavelengths 863 (lambdas), have only local significance. That is, a wavelength- 864 convertible network with full wavelength-conversion capability at 865 each node is equivalent to a circuit-switched TDM network with full 866 time slot interchange capability; thus, the routing problem needs to 867 be addressed only at the level of the traffic engineered (TE) link 868 choice, and wavelength assignment can be resolved locally by the 869 switches on a hop-by-hop basis. 871 However, in the limiting case of an optical network with no 872 wavelength converters, a light path (optical channel - OCh -) needs a 873 route from source to destination and must pick a single wavelength 874 that can be used along that path without "colliding" with the 875 wavelength used by any other light path that may share an optical 876 fiber. This is sometimes referred to as a "wavelength continuity 877 constraint". To ease up on this constraint while keeping network 878 costs in check a limited number of wavelength converters maybe 879 introduce at key points in the network [Chu03]. 881 In the general case of limited or no wavelength converters this 882 computation is known as the Routing and Wavelength Assignment (RWA) 883 problem [HZang00]. The "hardness" of this problem is well documented. 884 There, however, exist a number of reasonable approximate methods for 885 its solution [HZang00]. 887 The inputs to the basic RWA problem are the requested light paths 888 source and destination, the networks topology, the locations and 889 capabilities of any wavelength converters, and the wavelengths 890 available on each optical link. The output from an algorithm solving 891 the RWA problem is an explicit route through ROADMs, a wavelength for 892 the optical transmitter, and a set of locations (generally associated 893 with ROADMs or switches) where wavelength conversion is to occur and 894 the new wavelength to be used on each component link after that point 895 in the route. 897 It is to be noted that choice of specific RWA algorithm is out of the 898 scope for this document. However there are a number of different 899 approaches to dealing with the RWA algorithm that can affect the 900 division of effort between signaling, routing and PCE. 902 4.1. Architectural Approaches to RWA 904 Two general computational approaches are taken to solving the RWA 905 problem. Some algorithms utilize a two step procedure of path 906 selection followed by wavelength assignment, and others solve the 907 problem in a combined fashion. 909 In the following, three different ways of performing RWA in 910 conjunction with the control plane are considered. The choice of one 911 of these architectural approaches over another generally impacts the 912 demands placed on the various control plane protocols. 914 4.1.1. Combined RWA (R&WA) 916 In this case, a unique entity is in charge of performing routing and 917 wavelength assignment. This choice assumes that computational entity 918 has sufficient WSON network link/nodal information and topology to be 919 able to compute RWA. This solution relies on a sufficient knowledge 920 of network topology, of available network resources and of network 921 nodes capabilities. This knowledge has to be accessible to the entity 922 performing the routing and wavelength assignment. 924 This solution is compatible with most known RWA algorithms, and in 925 particular those concerned with network optimization. On the other 926 hand, this solution requires up-to-date and detailed network 927 information dissemination. 929 Such a computational entity could reside in two different logical 930 places: 932 o In a separate Path Computation Element (PCE) which hence owns the 933 complete and updated knowledge of network state and provides path 934 computation services to node. 936 o In the Ingress node, in that case all nodes have the R&WA 937 functionality; the knowledge of the network state is obtained by a 938 periodic flooding of information provided by the other nodes. 940 4.1.2. Separated R and WA (R+WA) 942 In this case a first entity performs routing, while a second performs 943 wavelength assignment. The first entity furnishes one or more paths 944 to the second entity that will perform wavelength assignment and 945 possibly final path selection. 947 As the entities computing the path and the wavelength assignment are 948 separated, this constrains the class of RWA algorithms that may be 949 implemented. Although it may seem that algorithms optimizing a joint 950 usage of the physical and spectral paths are excluded from this 951 solution, many practical optimization algorithms only consider a 952 limited set of possible paths, e.g., as computed via a k-shortest 953 path algorithm [Ozdaglar03]. Hence although there is no guarantee 954 that the selected final route and wavelength offers the optimal 955 solution by allowing multiple routes to pass to the wavelength 956 selection process reasonable optimization can be performed. 958 The entity performing the routing assignment needs the topology 959 information of the network, whereas the entity performing the 960 wavelength assignment needs information on the network's available 961 resources and on network node capabilities. 963 4.1.3. Routing and Distributed WA (R+DWA) 965 In this case a first entity performs routing, while wavelength 966 assignment is performed on a hop-by-hop manner along the previously 967 computed route. This mechanism relies on updating of a list of 968 potential wavelengths used to ensure conformance with the wavelength 969 continuity constraint. 971 As currently specified, the GMPLS protocol suite signaling protocol 972 can accommodate such an approach. Per [RFC3471], the Label Set 973 selection works according to an AND scheme. Each hop restricts the 974 Label Set sent to the next hop from the one received from the 975 previous hop by performing an AND operation between the wavelength 976 referred by the labels the message includes with the one available on 977 the ongoing interface. The constraint to perform this AND operation 978 is up to the node local policy (even if one expects a consistent 979 policy configuration throughout a given transparency domain). When 980 wavelength conversion is performed at an intermediate node, a new 981 Label Set is generated. The egress nodes selects one label in the 982 Label Set received at the node, which is also up to the node local 983 policy. 985 Depending on these policies a spectral assignment may not be found or 986 one consuming too many conversion resources relative to what a 987 dedicated wavelength assignment policy would have achieved. Hence, 988 this approach may generate higher blocking probabilities in a heavily 989 loaded network. 991 On the one hand, this solution may be empowered with some signaling 992 extensions to ease its functioning and possibly enhance its 993 performances relatively to blocking. Note that this approach requires 994 less information dissemination than the others. 996 The first entity may be a PCE or the ingress node of the LSP. This 997 solution is applicable inside networks where resource optimization is 998 not as critical. 1000 4.2. Conveying information needed by RWA 1002 The previous sections have characterized WSONs and lightpath 1003 requests. In particular, high level models of the information used by 1004 the RWA process were presented. We can view this information as 1005 either static, changing with hardware changes (including possibly 1006 failures), or dynamic, those that can change with subsequent 1007 lightpath provisioning. The timeliness in which an entity involved in 1008 the RWA process is notified of such changes is fairly situational. 1010 For example, for network restoration purposes, learning of a hardware 1011 failure or of new hardware coming online to provide restoration 1012 capability can be critical. 1013 Currently there are various methods for communicating RWA relevant 1014 information, these include, but are not limited to: 1016 o Existing control plane protocols such as GMPLS routing and 1017 signaling. Note that routing protocols can be used to convey both 1018 static and dynamic information. Static information currently 1019 conveyed includes items like router options and such. 1021 o Management protocols such as NetConf, SNMPv3, CLI, CORBA, or 1022 others. 1024 o Directory services and accompanying protocols. These are good for 1025 the dissemination of relatively static information. Not intended 1026 for dynamic information. 1028 o Other techniques for dynamic information: messaging straight from 1029 NEs to PCE to avoid flooding. This would be useful if the number 1030 of PCEs is significantly less than number of WSON NEs. Or other 1031 ways to limit flooding to "interested" NEs. 1033 Mechanisms to improve scaling of dynamic information: 1035 o Tailor message content to WSON. For example the use of wavelength 1036 ranges, or wavelength occupation bit maps. 1038 Utilize incremental updates if feasible. 1040 4.3. Lightpath Temporal Characteristics 1042 The temporal characteristics of a light path connection can affect 1043 the choice of solution to the RWA process. For our purposes here we 1044 look at the timeliness of connection establishment/teardown, and the 1045 duration of the connection. 1047 Connection Establishment/Teardown Timeliness can be thought of in 1048 approximately three time frames: 1050 1. Time Critical: For example those lightpath establishments used for 1051 restoration of service or other high priority real time service 1052 requests. 1054 2. Soft time bounds: This is a more typical new connection request. 1055 While expected to be responsive, there should be more time to take 1056 into account network optimization. 1058 3. Scheduled or Advanced reservations. Here lightpath connections are 1059 requested significantly ahead of their intended "in service" time. 1060 There is the potential for significant network optimization if 1061 multiple lightpaths can be computed concurrently to achieve network 1062 optimization objectives. 1064 Lightpath connection duration has typically been thought of as 1065 approximately three time frames: 1067 1. Dynamic: those lightpaths with relatively short duration (holding 1068 times). 1070 2. Pseudo-static: lightpaths with moderately long durations. 1072 3. Static: lightpaths with long durations. 1074 Different types of RWA algorithms have been developed for dealing 1075 with dynamic versus pseudo-static conditions. These can address 1076 service provider's needs for: (a) network optimization, (b) 1077 restoration, and (c) highly dynamic lightpath provisioning. 1079 Hence we can model timescale related lightpath requirements via the 1080 following notions: 1082 o Batch or Sequential light path connection requests 1084 o Timeliness of Connection establishment 1086 o Duration of lightpath connection 1088 5. Modeling Examples and Control Plane Use Cases 1090 This section provides examples of the fixed and switch optical node 1091 and wavelength constraint models of section 3. and WSON control plane 1092 use cases related to path computation, establishment, rerouting, and 1093 optimization. 1095 5.1. Network Modeling for GMPLS/PCE Control 1097 Consider a network containing three routers (R1 through R3), eight 1098 WSON nodes (N1 through N8) and 18 links (L1 through L18) and one OEO 1099 converter (O1) in a topology shown below. 1101 +--+ +--+ +--+ +--------+ 1102 +-L3-+N2+-L5-+ +--------L12--+N6+--L15--+ N8 +-- 1103 | +--+ |N4+-L8---+ +--+ ++--+---++ 1104 | | +-L9--+| | | | 1105 +--+ +-+-+ ++-+ || | L17 L18 1106 | ++-L1--+ | | ++++ +----L16---+ | | 1107 |R1| | N1| L7 |R2| | | | 1108 | ++-L2--+ | | ++-+ | ++---++ 1109 +--+ +-+-+ | | | + R3 | 1110 | +--+ ++-+ | | +-----+ 1111 +-L4-+N3+-L6-+N5+-L10-+ ++----+ 1112 +--+ | +--------L11--+ N7 +---- 1113 +--+ ++---++ 1114 | | 1115 L13 L14 1116 | | 1117 ++-+ | 1118 |O1+-+ 1119 +--+ 1120 5.1.1. Describing the WSON nodes 1122 The eight WSON nodes in this example have the following properties: 1124 o Nodes N1, N2, N3 have fixed OADMs (FOADMs) installed and can 1125 therefore only access a static and pre-defined set of wavelengths 1127 o All other nodes contain ROADMs and can therefore access all 1128 wavelengths. 1130 o Nodes N4, N5, N7 and N8 are multi-degree nodes, allowing any 1131 wavelength to be optically switched between any of the links. Note 1132 however, that this does not automatically apply to wavelengths 1133 that are being added or dropped at the particular node. 1135 o Node N4 is an exception to that: This node can switch any 1136 wavelength from its add/drop ports to any of its outgoing links 1137 (L5, L7 and L12 in this case) 1139 o The links from the routers are always only able to carry one 1140 wavelength with the exception of links L8 and L9 which are capable 1141 to add/drop any wavelength. 1143 o Node N7 contains an OEO transponder (O1) connected to the node via 1144 links L13 and L14. That transponder operates in 3R mode and does 1145 not change the wavelength of the signal. Assume that it can 1146 regenerate any of the client signals, however only for a specific 1147 wavelength. 1149 Given the above restrictions, the node information for the eight 1150 nodes can be expressed as follows: (where ID == identifier, SCM == 1151 switched connectivity matrix, and FCM == fixed connectivity matrix). 1153 +ID+SCM +FCM + 1154 | | |L1 |L2 |L3 |L4 | | |L1 |L2 |L3 |L4 | | 1155 | |L1 |0 |0 |0 |0 | |L1 |0 |0 |1 |0 | | 1156 |N1|L2 |0 |0 |0 |0 | |L2 |0 |0 |0 |1 | | 1157 | |L3 |0 |0 |0 |0 | |L3 |1 |0 |0 |1 | | 1158 | |L4 |0 |0 |0 |0 | |L4 |0 |1 |1 |0 | | 1159 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1160 | | |L3 |L5 | | | | |L3 |L5 | | | | 1161 |N2|L3 |0 |0 | | | |L3 |0 |1 | | | | 1162 | |L5 |0 |0 | | | |L5 |1 |0 | | | | 1163 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1164 | | |L4 |L6 | | | | |L4 |L6 | | | | 1165 |N3|L4 |0 |0 | | | |L4 |0 |1 | | | | 1166 | |L6 |0 |0 | | | |L6 |1 |0 | | | | 1167 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1168 | | |L5 |L7 |L8 |L9 |L12| |L5 |L7 |L8 |L9 |L12| 1169 | |L5 |0 |1 |1 |1 |1 |L5 |0 |0 |0 |0 |0 | 1170 |N4|L7 |1 |0 |1 |1 |1 |L7 |0 |0 |0 |0 |0 | 1171 | |L8 |1 |1 |0 |1 |1 |L8 |0 |0 |0 |0 |0 | 1172 | |L9 |1 |1 |1 |0 |1 |L9 |0 |0 |0 |0 |0 | 1173 | |L12|1 |1 |1 |1 |0 |L12|0 |0 |0 |0 |0 | 1174 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1175 | | |L6 |L7 |L10|L11| | |L6 |L7 |L10|L11| | 1176 | |L6 |0 |1 |0 |1 | |L6 |0 |0 |1 |0 | | 1177 |N5|L7 |1 |0 |0 |1 | |L7 |0 |0 |0 |0 | | 1178 | |L10|0 |0 |0 |0 | |L10|1 |0 |0 |0 | | 1179 | |L11|1 |1 |0 |0 | |L11|0 |0 |0 |0 | | 1180 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1181 | | |L12|L15| | | | |L12|L15| | | | 1182 |N6|L12|0 |1 | | | |L12|0 |0 | | | | 1183 | |L15|1 |0 | | | |L15|0 |0 | | | | 1184 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1185 | | |L11|L13|L14|L16| | |L11|L13|L14|L16| | 1186 | |L11|0 |1 |0 |1 | |L11|0 |0 |0 |0 | | 1187 |N7|L13|1 |0 |0 |0 | |L13|0 |0 |1 |0 | | 1188 | |L14|0 |0 |0 |1 | |L14|0 |1 |0 |0 | | 1189 | |L16|1 |0 |1 |0 | |L16|0 |0 |1 |0 | | 1190 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1191 | | |L15|L16|L17|L18| | |L15|L16|L17|L18| | 1192 | |L15|0 |1 |0 |0 | |L15|0 |0 |0 |1 | | 1193 |N8|L16|1 |0 |0 |0 | |L16|0 |0 |1 |0 | | 1194 | |L17|0 |0 |0 |0 | |L17|0 |1 |0 |0 | | 1195 | |L18|0 |0 |0 |0 | |L18|1 |0 |1 |0 | | 1196 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1198 5.1.2. Describing the links 1200 For the following discussion some simplifying assumptions are made: 1202 o It is assumed that the WSON node support a total of four 1203 wavelengths designated WL1 through WL4. 1205 o It is assumed that the impairment feasibility of a path or path 1206 segment is independent from the wavelength chosen. 1208 For the discussion of the RWA operation to build LSPs between two 1209 routers, the wavelength constraints on the links between the routers 1210 and the WSON nodes as well as the connectivity matrix of these links 1211 needs to be specified: 1213 +Link+WLs supported +Possible egress links+ 1214 | L1 | WL1 | L3 | 1215 +----+-----------------+---------------------+ 1216 | L2 | WL2 | L4 | 1217 +----+-----------------+---------------------+ 1218 | L8 | WL1 WL2 WL3 WL4 | L5 L7 L12 | 1219 +----+-----------------+---------------------+ 1220 | L9 | WL1 WL2 WL3 WL4 | L5 L7 L12 | 1221 +----+-----------------+---------------------+ 1222 | L10| WL2 | L6 | 1223 +----+-----------------+---------------------+ 1224 | L13| WL1 WL2 WL3 WL4 | L11 L14 | 1225 +----+-----------------+---------------------+ 1226 | L14| WL1 WL2 WL3 WL4 | L13 L16 | 1227 +----+-----------------+---------------------+ 1228 | L17| WL2 | L16 | 1229 +----+-----------------+---------------------+ 1230 | L18| WL1 | L15 | 1231 +----+-----------------+---------------------+ 1233 Note that the possible egress links for the links connecting to the 1234 routers is inferred from the Switched Connectivity Matrix and the 1235 Fixed Connectivity Matrix of the Nodes N1 through N8 and is show here 1236 for convenience, i.e., this information does not need to be repeated. 1238 5.2. RWA Path Computation and Establishment 1240 The calculation of optical impairment feasible routes is outside the 1241 scope of this framework document. In general impairment feasible 1242 routes serve as an input to the RWA algorithm. 1244 For the example use case shown here, assume the following feasible 1245 routes: 1247 +Endpoint 1+Endpoint 2+Feasible Route + 1248 | R1 | R2 | L1 L3 L5 L8 | 1249 | R1 | R2 | L1 L3 L5 L9 | 1250 | R1 | R2 | L2 L4 L6 L7 L8 | 1251 | R1 | R2 | L2 L4 L6 L7 L9 | 1252 | R1 | R2 | L2 L4 L6 L10 | 1253 | R1 | R3 | L1 L3 L5 L12 L15 L18 | 1254 | R1 | N7 | L2 L4 L6 L11 | 1255 | N7 | R3 | L16 L17 | 1256 | N7 | R2 | L16 L15 L12 L9 | 1257 | R2 | R3 | L8 L12 L15 L18 | 1258 | R2 | R3 | L8 L7 L11 L16 L17 | 1259 | R2 | R3 | L9 L12 L15 L18 | 1260 | R2 | R3 | L9 L7 L11 L16 L17 | 1262 Given a request to establish a LSP between R1 and R2 the RWA 1263 algorithm finds the following possible solutions: 1265 +WL + Path + 1266 | WL1| L1 L3 L5 L8 | 1267 | WL1| L1 L3 L5 L9 | 1268 | WL2| L2 L4 L6 L7 L8| 1269 | WL2| L2 L4 L6 L7 L9| 1270 | WL2| L2 L4 L6 L10 | 1272 Assume now that the RWA chooses WL1 and the Path L1 L3 L5 L8 for the 1273 requested LSP. 1275 Next, another LSP is signaled from R1 to R2. Given the established 1276 LSP using WL1, the following table shows the available paths: 1278 +WL + Path + 1279 | WL2| L2 L4 L6 L7 L9| 1280 | WL2| L2 L4 L6 L10 | 1282 Assume now that the RWA chooses WL2 and the path L2 L4 L6 L7 L9 for 1283 the establishment of the new LSP. 1285 Faced with another LSP request -this time from R2 to R3 - can not be 1286 fulfilled since the only four possible paths (starting at L8 and L9) 1287 are already in use. 1289 5.3. Resource Optimization 1291 The preceding example gives rise to another use case: The 1292 optimization of network resources. Optimization can be achieved on a 1293 number of layers (e.g. through electrical or optical multiplexing of 1294 client signals) or by re-optimizing the solutions found by the RWA 1295 algorithm. 1297 Given the above example again, assume that the RWA algorithm should 1298 find a path between R2 and R3. The only possible path to reach R3 1299 from R2 needs to use L9. L9 however is blocked by one of the LSPs 1300 from R1. 1302 5.4. Support for Rerouting 1304 It is also envisioned that the extensions to GMPLS and PCE support 1305 rerouting of wavelengths in case of failures. 1307 Assume for this discussion that the only two LSPs in use in the 1308 system are: 1310 LSP1: WL1 L1 L3 L5 L8 1312 LSP2: WL2 L2 L4 L6 L7 L9 1314 Assume furthermore that the link L5 fails. The RWA can now find the 1315 following alternate path and and establish that path: 1317 R1 -> N7 -> R2 1319 Level 3 regeneration will take place at N7, so that the complete path 1320 looks like this: 1322 R1 -> L2 L4 L6 L11 L13 -> O1 -> L14 L16 L15 L12 L9 -> R2 1324 6. GMPLS & PCE Implications 1326 The presence and amount of wavelength conversion available at a 1327 wavelength switching interface has an impact on the information that 1328 needs to be transferred by the control plane (GMPLS) and the PCE 1329 architecture. Current GMPLS and PCE standards can address the full 1330 wavelength conversion case so the following will only address the 1331 limited and no wavelength conversion cases. 1333 6.1. Implications for GMPLS signaling 1335 Basic support for WSON signaling already exists in GMPLS with the 1336 lambda (value 9) LSP encoding type [RFC3471], or for G.709 compatible 1337 optical channels, the LSP encoding type (value = 13) "G.709 Optical 1338 Channel" from [RFC4328]. However a number of practical issues arise 1339 in the identification of wavelengths and signals, and distributed 1340 wavelength assignment processes which are discussed below. 1342 6.1.1. Identifying Wavelengths and Signals 1344 As previously stated a global fixed mapping between wavelengths and 1345 labels simplifies the characterization of WDM links and WSON devices. 1346 Furthermore such a mapping as described in [Otani] eases 1347 communication between PCE and WSON PCCs. 1349 An alternative to a global network map of labels to wavelengths would 1350 be to use LMP to assign the map for each link then convey that 1351 information to any path computation entities, e.g., label switch 1352 routers or stand alone PCEs. The local label map approach will 1353 require the label-set contents in the RSVP-TE Path message to be 1354 translated every time the map changes between an incoming link and 1355 the outgoing link. 1357 In the future, it maybe worthwhile to define traffic parameters for 1358 lambda LSPs that include a signal type field that includes modulation 1359 format/rate information. This is similar to what was done in 1360 reference [RFC4606] for SONET/SDH signal types. 1362 6.1.2. Combined RWA/Separate Routing WA support 1364 In either the combined RWA or separate routing WA cases, the node 1365 initiating the signaling will have a route from the source to 1366 destination along with the wavelengths (generalized labels) to be 1367 used along portions of the path. Current GMPLS signaling supports an 1368 explicit route object (ERO) and within an ERO an ERO Label subobject 1369 can be use to indicate the wavelength to be used at a particular 1370 node. In case the local label map approach is used the label sub- 1371 object entry in the ERO has to be translated appropriately. 1373 6.1.3. Distributed Wavelength Assignment: Unidirectional, No 1374 Converters 1376 GMPLS signaling for a uni-directional lightpath LSP allows for the 1377 use of a label set object in the RSVP-TE path message. The processing 1378 of the label set object to take the intersection of available lambdas 1379 along a path can be performed resulting in the set of available 1380 lambda being known to the destination that can then use a wavelength 1381 selection algorithm to choose a lambda. For example, the following is 1382 a non-exhaustive subset of wavelength assignment (WA) approaches 1383 discussed in [HZang00]: 1385 1. Random: Looks at all available wavelengths for the light path then 1386 chooses from those available at random. 1388 2. First Fit: Wavelengths are ordered, first available (on all links) 1389 is chosen. 1391 3. Most Used: Out of the wavelengths available on the path attempts 1392 to select most use wavelength in network. 1394 4. Least Loaded: For multi-fiber networks. Chooses the wavelength j 1395 that maximizes minimum of the difference between the number of 1396 fibers on link l and the number of fibers on link l with 1397 wavelength j occupied. 1399 As can be seen from the above short list, wavelength assignment 1400 methods have differing information or processing requirements. The 1401 information requirements of these methods are as follows: 1403 1. Random: nothing more than the available wavelength set. 1405 2. First Fit: nothing more than the available wavelength set. 1407 3. Most Used: the available wavelength set and information on global 1408 wavelength use in the network. 1410 4. Least Loaded: the available wavelength set and information 1411 concerning the wavelength dependent loading for each link (this 1412 applies to multi-fiber links). This could be obtained via global 1413 information or via supplemental information passed via the 1414 signaling protocol. 1416 In case (3) above the global information needed by the wavelength 1417 assignment could be derived from suitably enhanced GMPLS routing. 1418 Note however this information need not be accurate enough for 1419 combined RWA computation. Currently, GMPLS signaling does not provide 1420 a way to indicate that a particular wavelength assignment algorithm 1421 should be used. 1423 6.1.4. Distributed Wavelength Assignment: Unidirectional, Limited 1424 Converters 1426 The previous outlined the case with no wavelength converters. In the 1427 case of wavelength converters, nodes with wavelength converters would 1428 need to make the decision as to whether to perform conversion. One 1429 indicator for this would be that the set of available wavelengths 1430 which is obtained via the intersection of the incoming label set and 1431 the egress links available wavelengths is either null or deemed too 1432 small to permit successful completion. 1434 At this point the node would need to remember that it will apply 1435 wavelength conversion and will be responsible for assigning the 1436 wavelength on the previous lambda-contiguous segment when the RSVP-TE 1437 RESV message passes by. The node will pass on an enlarged label set 1438 reflecting only the limitations of the wavelength converter and the 1439 egress link. The record route option in RVSP-TE signaling can be used 1440 to show where wavelength conversion has taken place. 1442 6.1.5. Distributed Wavelength Assignment: Bidirectional, No 1443 Converters 1445 There are potential issues in the case of a bi-directional lightpath 1446 which requires the use of the same lambda in both directions. We can 1447 try to use the above procedure to determine the available 1448 bidirectional lambda set if we use the interpretation that the 1449 available label set is available in both directions. However, a 1450 problem, arises in that bidirectional LSPs setup, according to 1451 [RFC3471] section 4.1, is indicated by the presence of an upstream 1452 label in the path message. 1454 However, until the intersection of the available label sets is 1455 obtained, e.g., at the destination node and the wavelength assignment 1456 algorithm has been run the upstream label information will not be 1457 available. Hence currently distributed wavelength assignment with 1458 bidirectional lightpaths is not supported. 1460 6.2. Implications for GMPLS Routing 1462 GMPLS routing [RFC4202] currently defines an interface capability 1463 descriptor for "lambda switch capable" (LSC) which we can use to 1464 describe the interfaces on a ROADM or other type of wavelength 1465 selective switch. In addition to the topology information typically 1466 conveyed via an IGP, we would need to convey the following subsystem 1467 properties to minimally characterize a WSON: 1469 1. WDM Link properties (allowed wavelengths). 1471 2. Laser Transmitters (wavelength range). 1473 3. ROADM/FOADM properties (connectivity matrix, port wavelength 1474 restrictions). 1476 4. Wavelength Converter properties (per network element, may change if 1477 a common limited shared pool is used). 1479 In most cases we should be able to combine items (1) and (2) into the 1480 information in item (3). Except for the number of wavelength 1481 converters that are available in a shared pool, and the previous 1482 information is fairly static. In the next two sections we discuss 1483 dynamic available link bandwidth information. 1485 6.2.1. Need for Wavelength-Specific Maximum Bandwidth Information 1487 Difficulties are encountered when trying to use the bandwidth 1488 accounting methods of [RFC4202] and [RFC3630] to describe the 1489 availability of wavelengths on a WDM link. The current RFCs give 1490 three link resource measures: Maximum Bandwidth, Maximum Reservable 1491 Bandwidth, and Unreserved Bandwidth. Although these can be used to 1492 describe a WDM span they do not provide the fundamental information 1493 needed for RWA. We are not given the maximum bandwidth per wavelength 1494 for the span. If we did then we could use the aforementioned measures 1495 to tell us the maximum wavelength count and the number of available 1496 wavelengths. 1498 For example, suppose we have a 32 channel WDM span, and that the 1499 system in general supports ITU-T NRZ signals up to NRZ 10Gbps. 1500 Further suppose that the first 20 channels are carrying 1Gbps 1501 Ethernet, then the maximum bandwidth would be 320Gbps and the maximum 1502 reservable bandwidth would be 120Gbps (12 wavelengths). 1503 Alternatively, consider the case where the first 8 channels are 1504 carrying 2.5Gbps SDH STM-16 channels, then the maximum bandwidth 1505 would still be 320Gbps and the maximum reservable bandwidth would be 1506 240Gbps (24 wavelengths). 1508 Such information would be useful in the routing with distributed WA 1509 approach of section 4.1.3. 1511 6.2.2. Need for Wavelength-Specific Availability Information 1513 Even if we know the number of available wavelengths on a link, we 1514 actually need to know which specific wavelengths are available and 1515 which are occupied if we are going to run a combined RWA process or 1516 separate WA process as discussed in sections 4.1.1. 4.1.2. This is 1517 currently not possible with GMPLS routing extensions. 1519 In the routing extensions for GMPLS [RFC4202], requirements for 1520 layer-specific TE attributes are discussed. The RWA problem for 1521 optical networks without wavelength converters imposes an additional 1522 requirement for the lambda (or optical channel) layer: that of 1523 knowing which specific wavelengths are in use. Note that current 1524 dense WDM (DWDM) systems range from 16 channels to 128 channels with 1525 advanced laboratory systems with as many as 300 channels. Given these 1526 channel limitations and if we take the approach of a global 1527 wavelength to label mapping or furnishing the local mappings to the 1528 PCEs then representing the use of wavelengths via a simple bit-map is 1529 feasible. 1531 6.2.3. Relationship to Link Bundling and Layering 1533 When dealing with static DWDM systems, particularly from a SONET/SDH 1534 or G.709 digital wrapper layer, each lambda looks like a separate 1535 link. Typically a bunch of unnumbered links, as supported in GMPLS 1536 routing extensions [RFC4202], would be used to describe a static DWDM 1537 system. In addition these links can be bundled into a TE link 1538 ([RFC4202], [RFC4201]) for more efficient dissemination of resource 1539 information. However, in the case discussed here we want to control a 1540 dynamic WDM layer and must deal with wavelengths as labels and not 1541 just as links or component links from the perspective of an upper 1542 (client) layer. In addition, a typical point to point optical cable 1543 contains many optical fibers and hence it may be desirable to bundle 1544 these separate fibers into a TE link. Note that in the no wavelength 1545 conversion or limited wavelength conversion situations that we will 1546 need information on wavelength usage on the individual component 1547 links. 1549 6.2.4. WSON Routing Information Summary 1551 The following table summarizes the WSON information that could be 1552 conveyed via GMPLS routing and attempts to classify that information 1553 as to its static or dynamic nature and whether that information would 1554 tend to be associated with either a link or a node. 1556 Information Static/Dynamic Node/Link 1557 ------------------------------------------------------------------ 1558 Connectivity matrix Static Node 1559 Per port wavelength restrictions Static Node(1) 1560 WDM link (fiber) lambda ranges Static Link 1561 WDM link channel spacing Static Link 1562 Laser Transmitter range Static Link(2) 1563 Wavelength conversion capabilities Static(3) Node 1564 Maximum bandwidth per Wavelength Static Link 1565 Wavelength Availability Dynamic(4) Link 1567 Notes: 1569 1. These are the per port wavelength restrictions of an optical 1570 device such as a ROADM and are independent of any optical 1571 constraints imposed by a fiber link. 1573 2. This could also be viewed as a node capability. 1575 3. This could be dynamic in the case of a limited pool of converters 1576 where the number available can change with connection 1577 establishment. Note we may want to include regeneration 1578 capabilities here since OEO converters are also regenerators. 1580 4. Not necessarily needed in the case of distributed wavelength 1581 assignment via signaling. 1583 While the full complement of the information from the previous table 1584 is needed in the Combined RWA and the separate Routing and WA 1585 architectures, in the case of Routing + distribute WA via signaling 1586 we only need the following information: 1588 Information Static/Dynamic Node/Link 1589 ------------------------------------------------------------------ 1590 Connectivity matrix Static Node 1591 Wavelength conversion capabilities Static(3) Node 1593 Information models and compact encodings for this information is 1594 provided in [WSON-Info]. 1596 6.3. Optical Path Computation and Implications for PCE 1598 As previously noted the RWA problem can be computationally intensive 1599 [HZang00]. Such computationally intensive path computations and 1600 optimizations were part of the impetus for the PCE (path computation 1601 element) architecture. 1603 As the PCEP defines the procedures necessary to support both 1604 sequential [PCEP] and global concurrent path computations [PCE-GCO], 1605 PCE is well positioned to support WSON-enabled RWA computation with 1606 some protocol enhancement. 1608 Implications for PCE generally fall into two main categories: (a) 1609 lightpath constraints and characteristics, (b) computation 1610 architectures. 1612 6.3.1. Lightpath Constraints and Characteristics 1614 For the varying degrees of optimization that may be encountered in a 1615 network the following models of bulk and sequential lightpath 1616 requests are encountered: 1618 o Batch optimization, multiple lightpaths requested at one time. 1620 o Lightpath(s) and backup lightpath(s) requested at one time. 1622 o Single lightpath requested at a time. 1624 PCEP and PCE-GCO can be readily enhanced to support all of the 1625 potential models of RWA computation. 1627 Lightpath constraints include: 1629 o Bidirectional Assignment of wavelengths 1631 o Possible simultaneous assignment of wavelength to primary and 1632 backup paths. 1634 o Tuning range constraint on optical transmitter. 1636 Lightpath characteristics can include: 1638 o Duration information (how long this connection may last) 1640 o Timeliness/Urgency information (how quickly is this connection 1641 needed) 1643 6.3.2. Computation Architecture Implications 1645 When a PCE performs a combined RWA computation per section 4.1.1. it 1646 requires accurate an up to date wavelength utilization on all links 1647 in the network. 1649 When a PCE is used to perform wavelength assignment (WA) in the 1650 separate routing WA architecture then the entity requesting WA needs 1651 to furnish the pre-selected route to the PCE as well as any of the 1652 lightpath constraints/characteristics previously mentioned. This 1653 architecture also requires the PCE performing WA to have accurate and 1654 up to date network wavelength utilization information. 1656 When a PCE is used to perform routing in a routing with distribute WA 1657 architecture, then the PCE does not necessarily need the most up to 1658 date network wavelength utilization information, however timely 1659 information can contributed to reducing failed signaling attempts 1660 related to blocking. 1662 6.3.3. Discovery of RWA Capable PCEs 1664 The algorithms and network information needed for solving the RWA are 1665 somewhat specialized and computationally intensive hence not all PCEs 1666 within a domain would necessarily need or want this capability. 1667 Hence, it would be useful via the mechanisms being established for 1668 PCE discovery [RFC5088] to indicate that a PCE has the ability to 1669 deal with the RWA problem. Reference [RFC5088] indicates that a sub- 1670 TLV could be allocated for this purpose. 1672 Recent progress on objective functions in PCE [PCE-OF] would allow 1673 the operators to flexibly request differing objective functions per 1674 their need and applications. For instance, this would allow the 1675 operator to choose an objective function that minimizes the total 1676 network cost associated with setting up a set of paths concurrently. 1677 This would also allow operators to choose an objective function that 1678 results in a most evenly distributed link utilization. 1680 This implies that PCEP would easily accommodate wavelength selection 1681 algorithm in its objective function to be able to optimize the path 1682 computation from the perspective of wavelength assignment if chosen 1683 by the operators. 1685 6.4. Scaling Implications 1687 This section provides a summary of the scaling issue for WSON 1688 routing, signaling and path computation introduced by the concepts 1689 discussed in this document. 1691 6.4.1. Routing 1693 In large WSONs label availability and cross connect capability 1694 information being advertised may generate a significant amount of 1695 routing information. 1697 6.4.2. Signaling 1699 When dealing with a large number of simultaneous end-to-end 1700 wavelength service requests and service deletions the network may 1701 have to process a significant number of forward and backward service 1702 messages. Also, similar situation possibly happens in the case of 1703 link or node failure, if the WSON support dynamic restoration 1704 capability. 1706 6.4.3. Path computation 1708 If a PCE is handling path computation requests for end-to-end 1709 wavelength services within the WSON, then the complexity of the 1710 network and number of service path computation requests being sent to 1711 the PCE may have an impact on the PCEs ability to process requests in 1712 a timely manner. 1714 6.5. Summary of Impacts by RWA Architecture 1716 The following table summarizes for each RWA strategy the list of 1717 mandatory ("M") and optional ("O") control plane features according 1718 to GMPLS architectural blocks: 1720 o Information required by the path computation entity, 1722 o LSP request parameters used in either PCC to PCE situations or in 1723 signaling, 1725 o RSVP-TE LSP signaling parameters used in LSP establishment. 1727 The table shows which enhancements are common to all architectures 1728 (R&WA, R+WA, R+DWA), which apply only to R&WA and R+WA (R+&WA), and 1729 which apply only to R+DWA. 1731 +-------------------------------------+-----+-------+-------+-------+ 1732 | | |Common | R+&WA | R+DWA | 1733 | Feature | ref +---+---+---+---+---+---+ 1734 | | | M | O | M | O | M | O | 1735 +-------------------------------------+-----+---+---+---+---+---+---+ 1736 | Generalized Label for Wavelength |5.1.1| x | | | | | | 1737 +-------------------------------------+-----+---+---+---+---+---+---+ 1738 | Flooding of information for the | | | | | | | | 1739 | routing phase | | | | | | | | 1740 | Node features | 3.3 | | | | | | | 1741 | Node type | | | x | | | | | 1742 | spectral X-connect constraint | | | | x | | | | 1743 | port X-connect constraint | | | | x | | | | 1744 | Transponders availability | | | x | | | | | 1745 | Transponders features | 3.2 | | x | | | | | 1746 | Converter availability | | | | x | | | | 1747 | Converter features | 3.4 | | | x | | | x | 1748 | TE-parameters of WDM links | 3.1 | x | | | | | | 1749 | Total Number of wavelength | | x | | | | | | 1750 | Number of wavelengths available | | x | | | | | | 1751 | Grid spacing | | x | | | | | | 1752 | Wavelength availability on links | 5.2 | | | x | | | | 1753 +-------------------------------------+-----+---+---+---+---+---+---+ 1754 | LSP request parameters | | | | | | | | 1755 | Signal features | 5.1 | | x | | | x | | 1756 | Modulation format | | | x | | | x | | 1757 | Modulation parameters | | | x | | | x | | 1758 | Specification of RWA method | 5.1 | | x | | | x | | 1759 | LSP time features | 4.3 | | x | | | | | 1760 +-------------------------------------+-----+---+---+---+---+---+---+ 1761 | Enriching signaling messages | | | | | | | | 1762 | Signal features | 5.1 | | | | | x | | 1763 +-------------------------------------+-----+---+---+---+---+---+---+ 1765 7. Security Considerations 1767 This document has no requirement for a change to the security models 1768 within GMPLS and associated protocols. That is the OSPF-TE, RSVP-TE, 1769 and PCEP security models could be operated unchanged. 1771 However satisfying the requirements for RWA using the existing 1772 protocols may significantly affect the loading of those protocols. 1773 This makes the operation of the network more vulnerable to denial of 1774 service attacks. Therefore additional care maybe required to ensure 1775 that the protocols are secure in the WSON environment. 1777 Furthermore the additional information distributed in order to 1778 address the RWA problem represents a disclosure of network 1779 capabilities that an operator may wish to keep private. Consideration 1780 should be given to securing this information. 1782 8. IANA Considerations 1784 This document makes no request for IANA actions. 1786 9. Acknowledgments 1788 The authors would like to thank Adrian Farrel for many helpful 1789 comments that greatly improved the contents of this draft. 1791 This document was prepared using 2-Word-v2.0.template.dot. 1793 10. References 1795 10.1. Normative References 1797 [RFC3471] Berger, L., "Generalized Multi-Protocol Label Switching 1798 (GMPLS) Signaling Functional Description", RFC 3471, 1799 January 2003. 1801 [RFC3630] Katz, D., Kompella, K., and D. Yeung, "Traffic Engineering 1802 (TE) Extensions to OSPF Version 2", RFC 3630, September 1803 2003. 1805 [RFC3945] Mannie, E., "Generalized Multi-Protocol Label Switching 1806 (GMPLS) Architecture", RFC 3945, October 2004. 1808 [RFC4201] Kompella, K., Rekhter, Y., and L. Berger, "Link Bundling in 1809 MPLS Traffic Engineering (TE)", RFC 4201, October 2005. 1811 [RFC4202] Kompella, K. and Y. Rekhter, "Routing Extensions in Support 1812 of Generalized Multi-Protocol Label Switching (GMPLS)", RFC 1813 4202, October 2005. 1815 [RFC4328] Papadimitriou, D., "Generalized Multi-Protocol Label 1816 Switching (GMPLS) Signaling Extensions for G.709 Optical 1817 Transport Networks Control", RFC 4328, January 2006. 1819 [G.694.1] ITU-T Recommendation G.694.1, "Spectral grids for WDM 1820 applications: DWDM frequency grid", June, 2002. 1822 [RFC5088] J.L. Le Roux, J.P. Vasseur, Yuichi Ikejiri, and Raymond 1823 Zhang, "OSPF protocol extensions for Path Computation 1824 Element (PCE) Discovery", January 2008. 1826 [PCE-GCO] Y. Lee, J.L. Le Roux, D. King, and E. Oki, "Path 1827 Computation Element Communication Protocol (PCECP) 1828 Requirements and Protocol Extensions In Support of Global 1829 Concurrent Optimization", work in progress, draft-ietf-pce- 1830 global-concurrent-optimization-08.txt, January 2009. 1832 [PCEP] J.P. Vasseur and J.L. Le Roux (Editors), "Path Computation 1833 Element (PCE) Communication Protocol (PCEP)", work in 1834 progress, draft-ietf-pce-pcep-19.txt, November 2008. 1836 [PCE-OF] J.L. Le Roux, J.P. Vasseur, and Y. Lee, "Encoding of 1837 Objective Functions in Path Computation Element (PCE) 1838 communication and discovery protocols", work in progress, 1839 draft-ietf-pce-of-06.txt, December 2008. 1841 [WSON-Encode] G. Bernstein, Y. Lee, D. Li, and W. Imajuku, "Routing 1842 and Wavelength Assignment Information Encoding for 1843 Wavelength Switched Optical Networks", draft-bernstein- 1844 ccamp-wson-encode-01.txt, November 2008. 1846 [WSON-Info] G. Bernstein, Y. Lee, D. Li, W. Imajuku," Routing and 1847 Wavelength Assignment Information for Wavelength Switched 1848 Optical Networks", draft-bernstein-ccamp-wson-info-03.txt, 1849 July, 2008. 1851 10.2. Informative References 1853 [HZang00] H. Zang, J. Jue and B. Mukherjeee, "A review of routing and 1854 wavelength assignment approaches for wavelength-routed 1855 optical WDM networks", Optical Networks Magazine, January 1856 2000. 1858 [Coldren04] Larry A. Coldren, G. A. Fish, Y. Akulova, J. S. 1859 Barton, L. Johansson and C. W. Coldren, "Tunable 1860 Seiconductor Lasers: A Tutorial", Journal of Lightwave 1861 Technology, vol. 22, no. 1, pp. 193-202, January 2004. 1863 [Chu03] Xiaowen Chu, Bo Li and Chlamtac I, "Wavelength converter 1864 placement under different RWA algorithms in wavelength- 1865 routed all-optical networks", IEEE Transactions on 1866 Communications, vol. 51, no. 4, pp. 607-617, April 2003. 1868 [Buus06] Jens Buus EJM, "Tunable Lasers in Optical Networks", 1869 Journal of Lightware Technology, vol. 24, no. 1, pp. 5-11, 1870 January 2006. 1872 [Basch06] E. Bert Bash, Roman Egorov, Steven Gringeri and Stuart 1873 Elby, "Architectural Tradeoffs for Reconfigurable Dense 1874 Wavelength-Division Multiplexing Systems", IEEE Journal of 1875 Selected Topics in Quantum Electronics, vol. 12, no. 4, pp. 1876 615-626, July/August 2006. 1878 [Otani] T. Otani, H. Guo, K. Miyazaki, D. Caviglia, "Generalized 1879 Labels of Lambda-Switching Capable Label Switching Routers 1880 (LSR)", work in progress: draft-otani-ccamp-gmpls-lambda- 1881 labels-02.txt, November 2007. 1883 [Winzer06] Peter J. Winzer and Rene-Jean Essiambre, "Advanced 1884 Optical Modulation Formats", Proceedings of the IEEE, vol. 1885 94, no. 5, pp. 952-985, May 2006. 1887 [G.652] ITU-T Recommendation G.652, Characteristics of a single-mode 1888 optical fibre and cable, June 2005. 1890 [G.653] ITU-T Recommendation G.653, Characteristics of a dispersion- 1891 shifted single-mode optical fibre and cable, December 2006. 1893 [G.654] ITU-T Recommendation G.654, Characteristics of a cut-off 1894 shifted single-mode optical fibre and cable, December 2006. 1896 [G.655] ITU-T Recommendation G.655, Characteristics of a non-zero 1897 dispersion-shifted single-mode optical fibre and cable, 1898 March 2006. 1900 [G.656] ITU-T Recommendation G.656, Characteristics of a fibre and 1901 cable with non-zero dispersion for wideband optical 1902 transport, December 2006. 1904 [G.671] ITU-T Recommendation G.671, Transmission characteristics of 1905 optical components and subsystems, January 2005. 1907 [G.872] ITU-T Recommendation G.872, Architecture of optical 1908 transport networks, November 2001. 1910 [G.959.1] ITU-T Recommendation G.959.1, Optical Transport Network 1911 Physical Layer Interfaces, March 2006. 1913 [G.694.1] ITU-T Recommendation G.694.1, Spectral grids for WDM 1914 applications: DWDM frequency grid, June 2002. 1916 [G.694.2] ITU-T Recommendation G.694.2, Spectral grids for WDM 1917 applications: CWDM wavelength grid, December 2003. 1919 [G.Sup39] ITU-T Series G Supplement 39, Optical system design and 1920 engineering considerations, February 2006. 1922 [G.Sup43] ITU-T Series G Supplement 43, Transport of IEEE 10G base-R 1923 in optical transport networks (OTN), November 2006. 1925 [Imajuku] W. Imajuku, Y. Sone, I. Nishioka, S. Seno, "Routing 1926 Extensions to Support Network Elements with Switching 1927 Constraint", work in progress: draft-imajuku-ccamp-rtg- 1928 switching-constraint-02.txt, July 2007. 1930 [Ozdaglar03] Asuman E. Ozdaglar and Dimitri P. Bertsekas, ''Routing 1931 and wavelength assignment in optical networks,'' IEEE/ACM 1932 Transactions on Networking, vol. 11, 2003, pp. 259 -272. 1934 [RFC4054] Strand, J. and A. Chiu, "Impairments and Other Constraints 1935 on Optical Layer Routing", RFC 4054, May 2005. 1937 [RFC4606] Mannie, E. and D. Papadimitriou, "Generalized Multi- 1938 Protocol Label Switching (GMPLS) Extensions for Synchronous 1939 Optical Network (SONET) and Synchronous Digital Hierarchy 1940 (SDH) Control", RFC 4606, August 2006. 1942 [WC-Pool] G. Bernstein, Y. Lee, "Modeling WDM Switching Systems 1943 including Wavelength Converters" to appear www.grotto- 1944 networking.com, 2008. 1946 11. Contributors 1948 Snigdho Bardalai 1949 Fujitsu 1950 Email: Snigdho.Bardalai@us.fujitsu.com 1952 Diego Caviglia 1953 Ericsson 1954 Via A. Negrone 1/A 16153 1955 Genoa Italy 1957 Phone: +39 010 600 3736 1958 Email: diego.caviglia@(marconi.com, ericsson.com) 1960 Daniel King 1961 Aria Networks 1962 Email: daniel.king@aria-networks.com 1964 Itaru Nishioka 1965 NEC Corp. 1966 1753 Simonumabe, Nakahara-ku, Kawasaki, Kanagawa 211-8666 1967 Japan 1968 Phone: +81 44 396 3287 1969 Email: i-nishioka@cb.jp.nec.com 1971 Lyndon Ong 1972 Ciena 1973 Email: Lyong@Ciena.com 1975 Pierre Peloso 1976 Alcatel-Lucent 1977 Route de Villejust - - 91620 Nozay - France 1978 Email: pierre.peloso@alcatel-lucent.fr 1980 Jonathan Sadler 1981 Tellabs 1982 Email: Jonathan.Sadler@tellabs.com 1984 Dirk Schroetter 1985 Cisco 1986 Email: dschroet@cisco.com 1988 Author's Addresses 1990 Greg M. Bernstein (ed.) 1991 Grotto Networking 1992 Fremont California, USA 1993 Phone: (510) 573-2237 1994 Email: gregb@grotto-networking.com 1996 Young Lee (ed.) 1997 Huawei Technologies 1998 1700 Alma Drive, Suite 100 1999 Plano, TX 75075 2000 USA 2002 Phone: (972) 509-5599 (x2240) 2003 Email: ylee@huawei.com 2005 Wataru Imajuku 2006 NTT Network Innovation Labs 2007 1-1 Hikari-no-oka, Yokosuka, Kanagawa 2008 Japan 2010 Phone: +81-(46) 859-4315 2011 Email: imajuku.wataru@lab.ntt.co.jp 2013 Intellectual Property Statement 2015 The IETF Trust takes no position regarding the validity or scope of 2016 any Intellectual Property Rights or other rights that might be 2017 claimed to pertain to the implementation or use of the technology 2018 described in any IETF Document or the extent to which any license 2019 under such rights might or might not be available; nor does it 2020 represent that it has made any independent effort to identify any 2021 such rights. 2023 Copies of Intellectual Property disclosures made to the IETF 2024 Secretariat and any assurances of licenses to be made available, or 2025 the result of an attempt made to obtain a general license or 2026 permission for the use of such proprietary rights by implementers or 2027 users of this specification can be obtained from the IETF on-line IPR 2028 repository at http://www.ietf.org/ipr 2030 The IETF invites any interested party to bring to its attention any 2031 copyrights, patents or patent applications, or other proprietary 2032 rights that may cover technology that may be required to implement 2033 any standard or specification contained in an IETF Document. Please 2034 address the information to the IETF at ietf-ipr@ietf.org. 2036 Disclaimer of Validity 2038 All IETF Documents and the information contained therein are provided 2039 on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE 2040 REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE 2041 IETF TRUST AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL 2042 WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY 2043 WARRANTY THAT THE USE OF THE INFORMATION THEREIN WILL NOT INFRINGE 2044 ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS 2045 FOR A PARTICULAR PURPOSE. 2047 Acknowledgment 2049 Funding for the RFC Editor function is currently provided by the 2050 Internet Society.