idnits 2.17.1 draft-ietf-ccamp-rwa-wson-framework-12.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (February 8, 2011) is 4798 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 ---------------------------------------------------------------------------- == Missing Reference: 'G.698.1' is mentioned on line 461, but not defined == Missing Reference: 'G.698.2' is mentioned on line 462, but not defined == Missing Reference: 'G.707' is mentioned on line 486, but not defined == Missing Reference: 'G.709' is mentioned on line 487, but not defined == Unused Reference: 'G.Sup43' is defined on line 2145, but no explicit reference was found in the text Summary: 0 errors (**), 0 flaws (~~), 6 warnings (==), 2 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: August 2011 Grotto Networking 5 Wataru Imajuku 6 NTT 8 February 8, 2011 10 Framework for GMPLS and PCE Control of Wavelength Switched Optical 11 Networks (WSON) 12 draft-ietf-ccamp-rwa-wson-framework-12.txt 14 Abstract 16 This document provides a framework for applying Generalized Multi- 17 Protocol Label Switching (GMPLS) and the Path Computation Element 18 (PCE) architecture to the control of wavelength switched optical 19 networks (WSON). In particular, it examines Routing and Wavelength 20 Assignment (RWA) of optical paths. 22 This document focuses on topological elements and path selection 23 constraints that are common across different WSON environments as 24 such it does not address optical impairments in any depth. 26 Status of this Memo 28 This Internet-Draft is submitted to IETF in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF), its areas, and its working groups. Note that 33 other groups may also distribute working documents as Internet- 34 Drafts. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 The list of current Internet-Drafts can be accessed at 42 http://www.ietf.org/ietf/1id-abstracts.txt 43 The list of Internet-Draft Shadow Directories can be accessed at 44 http://www.ietf.org/shadow.html. 46 This Internet-Draft will expire on August 8, 2011. 48 Copyright Notice 50 Copyright (c) 2011 IETF Trust and the persons identified as the 51 document authors. All rights reserved. 53 This document is subject to BCP 78 and the IETF Trust's Legal 54 Provisions Relating to IETF Documents 55 (http://trustee.ietf.org/license-info) in effect on the date of 56 publication of this document. Please review these documents 57 carefully, as they describe your rights and restrictions with 58 respect to this document. Code Components extracted from this 59 document must include Simplified BSD License text as described in 60 Section 4.e of the Trust Legal Provisions and are provided without 61 warranty as described in the Simplified BSD License. 63 Table of Contents 65 1. Introduction...................................................4 66 2. Terminology....................................................5 67 3. Wavelength Switched Optical Networks...........................6 68 3.1. WDM and CWDM Links........................................6 69 3.2. Optical Transmitters and Receivers........................8 70 3.3. Optical Signals in WSONs..................................9 71 3.3.1. Optical Tributary Signals...........................10 72 3.3.2. WSON Signal Characteristics.........................10 73 3.4. ROADMs, OXCs, Splitters, Combiners and FOADMs............11 74 3.4.1. Reconfigurable Add/Drop Multiplexers and OXCs.......11 75 3.4.2. Splitters...........................................14 76 3.4.3. Combiners...........................................15 77 3.4.4. Fixed Optical Add/Drop Multiplexers.................15 78 3.5. Electro-Optical Systems..................................16 79 3.5.1. Regenerators........................................16 80 3.5.2. OEO Switches........................................19 81 3.6. Wavelength Converters....................................19 82 3.6.1. Wavelength Converter Pool Modeling..................21 83 3.7. Characterizing Electro-Optical Network Elements..........25 84 3.7.1. Input Constraints...................................26 85 3.7.2. Output Constraints..................................26 86 3.7.3. Processing Capabilities.............................27 87 4. Routing and Wavelength Assignment and the Control Plane.......28 88 4.1. Architectural Approaches to RWA..........................28 89 4.1.1. Combined RWA (R&WA).................................29 90 4.1.2. Separated R and WA (R+WA)...........................29 91 4.1.3. Routing and Distributed WA (R+DWA)..................30 92 4.2. Conveying information needed by RWA......................30 93 5. Modeling Examples and Control Plane Use Cases.................31 94 5.1. Network Modeling for GMPLS/PCE Control...................31 95 5.1.1. Describing the WSON nodes...........................32 96 5.1.2. Describing the links................................34 97 5.2. RWA Path Computation and Establishment...................35 98 5.3. Resource Optimization....................................36 99 5.4. Support for Rerouting....................................37 100 5.5. Electro-Optical Networking Scenarios.....................37 101 5.5.1. Fixed Regeneration Points...........................37 102 5.5.2. Shared Regeneration Pools...........................38 103 5.5.3. Reconfigurable Regenerators.........................38 104 5.5.4. Relation to Translucent Networks....................38 105 6. GMPLS and PCE Implications....................................39 106 6.1. Implications for GMPLS signaling.........................39 107 6.1.1. Identifying Wavelengths and Signals.................39 108 6.1.2. WSON Signals and Network Element Processing.........40 109 6.1.3. Combined RWA/Separate Routing WA support............40 110 6.1.4. Distributed Wavelength Assignment: Unidirectional, No 111 Converters.................................................41 112 6.1.5. Distributed Wavelength Assignment: Unidirectional, 113 Limited Converters.........................................41 114 6.1.6. Distributed Wavelength Assignment: Bidirectional, No 115 Converters.................................................41 116 6.2. Implications for GMPLS Routing...........................42 117 6.2.1. Electro-Optical Element Signal Compatibility........42 118 6.2.2. Wavelength-Specific Availability Information........43 119 6.2.3. WSON Routing Information Summary....................43 120 6.3. Optical Path Computation and Implications for PCE........45 121 6.3.1. Optical path Constraints and Characteristics........45 122 6.3.2. Electro-Optical Element Signal Compatibility........45 123 6.3.3. Discovery of RWA Capable PCEs.......................46 124 7. Security Considerations.......................................46 125 8. IANA Considerations...........................................47 126 9. Acknowledgments...............................................47 127 10. References...................................................48 128 10.1. Normative References....................................48 129 10.2. Informative References..................................49 130 11. Contributors.................................................51 131 Author's Addresses...............................................52 132 Intellectual Property Statement..................................52 133 Disclaimer of Validity...........................................53 135 1. Introduction 137 Wavelength Switched Optical Networks (WSONs) are constructed from 138 subsystems that include Wavelength Division Multiplexed (WDM) links, 139 tunable transmitters and receivers, Reconfigurable Optical Add/Drop 140 Multiplexers (ROADM), wavelength converters, and electro-optical 141 network elements. A WSON is a WDM-based optical network in which 142 switching is performed selectively based on the center wavelength of 143 an optical signal. 145 WSONs can differ from other types of GMPLS networks in that many 146 types of WSON nodes are highly asymmetric with respect to their 147 switching capabilities, compatibility of signal types and network 148 elements may need to be considered, and label assignment can be non- 149 local. In order to provision an optical connection (an optical path) 150 through a WSON certain wavelength continuity and resource 151 availability constraints must be met to determine viable and optimal 152 paths through the WSON. The determination of paths is known as 153 Routing and Wavelength Assignment (RWA). 155 Generalized Multi-Protocol Label Switching (GMPLS) [RFC3945] includes 156 an architecture and a set of control plane protocols that can be used 157 to operate data networks ranging from packet switch capable networks, 158 through those networks that use time division multiplexing, to WDM 159 networks. The Path Computation Element (PCE) architecture [RFC4655] 160 defines functional components that can be used to compute and suggest 161 appropriate paths in connection-oriented traffic-engineered networks. 163 This document provides a framework for applying the GMPLS 164 architecture and protocols [RFC3945], and the PCE architecture 165 [RFC4655] to the control and operation of WSONs. To aid in this 166 process this document also provides an overview of the subsystems and 167 processes that comprise WSONs, and describes RWA so that the 168 information requirements, both static and dynamic, can be identified 169 to explain how the information can be modeled for use by GMPLS and 170 PCE systems. This work will facilitate the development of protocol 171 solution models and protocol extensions within the GMPLS and PCE 172 protocol families. 174 Different WSONs such as access, metro, and long haul may apply 175 different techniques for dealing with optical impairments hence this 176 document does not address optical impairments in any depth. Note that 177 this document focuses on the generic properties of links, switches 178 and path selection constraints that occur in many types of WSONs. 179 See [WSON-Imp] for more information on optical impairments and GMPLS. 181 2. Terminology 183 Add/Drop Multiplexers (ADM): An optical device used in WDM networks 184 composed of one or more line side ports and typically many tributary 185 ports. 187 CWDM: Coarse Wavelength Division Multiplexing. 189 DWDM: Dense Wavelength Division Multiplexing. 191 Degree: The degree of an optical device (e.g., ROADM) is given by a 192 count of its line side ports. 194 Drop and continue: A simple multi-cast feature of some ADM where a 195 selected wavelength can be switched out of both a tributary (drop) 196 port and a line side port. 198 FOADM: Fixed Optical Add/Drop Multiplexer. 200 GMPLS: Generalized Multi-Protocol Label Switching. 202 Line side: In WDM system line side ports and links typically can 203 carry the full multiplex of wavelength signals, as compared to 204 tributary (add or drop ports) that typically carry a few (typically 205 one) wavelength signals. 207 OXC: Optical cross connect. An optical switching element in which a 208 signal on any input port can reach any output port. 210 PCC: Path Computation Client. Any client application requesting a 211 path computation to be performed by the Path Computation Element. 213 PCE: Path Computation Element. An entity (component, application, or 214 network node) that is capable of computing a network path or route 215 based on a network graph and applying computational constraints. 217 PCEP: PCE Communication Protocol. The communication protocol between 218 a Path Computation Client and Path Computation Element. 220 ROADM: Reconfigurable Optical Add/Drop Multiplexer. A wavelength 221 selective switching element featuring input and output line side 222 ports as well as add/drop tributary ports. 224 RWA: Routing and Wavelength Assignment. 226 Transparent Network: A wavelength switched optical network that does 227 not contain regenerators or wavelength converters. 229 Translucent Network: A wavelength switched optical network that is 230 predominantly transparent but may also contain limited numbers of 231 regenerators and/or wavelength converters. 233 Tributary: A link or port on a WDM system that can carry 234 significantly less than the full multiplex of wavelength signals 235 found on the line side links/ports. Typical tributary ports are the 236 add and drop ports on an ADM and these support only a single 237 wavelength channel. 239 Wavelength Conversion/Converters: The process of converting an 240 information bearing optical signal centered at a given wavelength to 241 one with "equivalent" content centered at a different wavelength. 242 Wavelength conversion can be implemented via an optical-electronic- 243 optical (OEO) process or via a strictly optical process. 245 WDM: Wavelength Division Multiplexing. 247 Wavelength Switched Optical Networks (WSONs): WDM based optical 248 networks in which switching is performed selectively based on the 249 center wavelength of an optical signal. 251 3. Wavelength Switched Optical Networks 253 WSONs range in size from continent spanning long haul networks, to 254 metropolitan networks, to residential access networks. In all these 255 cases, the main concern is those properties that constrain the choice 256 of wavelengths that can be used, i.e., restrict the wavelength label 257 set, impact the path selection process, and limit the topological 258 connectivity. In addition, if electro-optical network elements are 259 used in the WSON, additional compatibility constraints may be imposed 260 by the network elements on various optical signal parameters. The 261 subsequent sections review and model some of the major subsystems of 262 a WSON with an emphasis on those aspects that are of relevance to the 263 control plane. In particular, WDM links, optical transmitters, 264 ROADMs, and wavelength converters are examined. 266 3.1. WDM and CWDM Links 268 WDM and CWDM links run over optical fibers, and optical fibers come 269 in a wide range of types that tend to be optimized for various 270 applications. Examples include access networks, metro, long haul, and 271 submarine links. International Telecommunication Union - 272 Telecommunication Standardization Sector (ITU-T) standards exist for 273 various types of fibers. Although fiber can be categorized into 274 Single mode fibers (SMF) and Multi-mode fibers (MMF), the latter are 275 typically used for short-reach campus and premise applications. SMF 276 are used for longer-reach applications and therefore are the primary 277 concern of this document. The following SMF fiber types are typically 278 encountered in optical networks: 280 ITU-T Standard | Common Name 281 ------------------------------------------------------------ 282 G.652 [G.652] | Standard SMF | 283 G.653 [G.653] | Dispersion shifted SMF | 284 G.654 [G.654] | Cut-off shifted SMF | 285 G.655 [G.655] | Non-zero dispersion shifted SMF | 286 G.656 [G.656] | Wideband non-zero dispersion shifted SMF | 287 ------------------------------------------------------------ 289 Typically WDM links operate in one or more of the approximately 290 defined optical bands [G.Sup39]: 292 Band Range (nm) Common Name Raw Bandwidth (THz) 293 O-band 1260-1360 Original 17.5 294 E-band 1360-1460 Extended 15.1 295 S-band 1460-1530 Short 9.4 296 C-band 1530-1565 Conventional 4.4 297 L-band 1565-1625 Long 7.1 298 U-band 1625-1675 Ultra-long 5.5 300 Not all of a band may be usable, for example in many fibers that 301 support E-band there is significant attenuation due to a water 302 absorption peak at 1383nm. Hence a discontinuous acceptable 303 wavelength range for a particular link may be needed and is modeled. 304 Also some systems will utilize more than one band. This is 305 particularly true for CWDM systems. 307 Current technology subdivides the bandwidth capacity of fibers into 308 distinct channels based on either wavelength or frequency. There are 309 two standards covering wavelengths and channel spacing. ITU-T 310 Recommendation G.694.1, Spectral grids for WDM applications: DWDM 311 frequency grid [G.694.1] describes a DWDM grid defined in terms of 312 frequency grids of 12.5GHz, 25GHz, 50GHz, 100GHz, and other multiples 313 of 100GHz around a 193.1THz center frequency. At the narrowest 314 channel spacing this provides less than 4800 channels across the O 315 through U bands. ITU-T Recommendation G.694.2, Spectral grids for WDM 316 applications: CWDM wavelength grid [G.694.2] describes a CWDM grid 317 defined in terms of wavelength increments of 20nm running from 1271nm 318 to 1611nm for 18 or so channels. The number of channels is 319 significantly smaller than the 32 bit GMPLS label space defined for 320 GMPLS, see [RFC3471]. A label representation for these ITU-T grids 321 is given in [Otani] and provides a common label format to be used in 322 signaling optical paths. Further, these ITU-T grid based labels can 323 also be used to describe WDM links, ROADM ports, and wavelength 324 converters for the purposes of path selection. 326 Many WDM links are designed to take advantage of particular fiber 327 characteristics or to try to avoid undesirable properties. For 328 example dispersion shifted SMF [G.653] was originally designed for 329 good long distance performance in single channel systems, however 330 putting WDM over this type of fiber requires significant system 331 engineering and a fairly limited range of wavelengths. Hence the 332 following information is needed as parameters to perform basic, 333 impairment unaware, modeling of a WDM link: 335 o Wavelength range(s): Given a mapping between labels and the ITU-T 336 grids each range could be expressed in terms of a tuple (lambda1, 337 lambda2) or (freq1, freq1) where the lambdas or frequencies can be 338 represented by 32 bit integers. 340 o Channel spacing: Currently there are five channel spacings used in 341 DWDM systems and a single channel spacing defined for CWDM 342 systems. 344 For a particular link this information is relatively static, as 345 changes to these properties generally require hardware upgrades. Such 346 information may be used locally during wavelength assignment via 347 signaling, similar to label restrictions in MPLS or used by a PCE in 348 providing combined RWA. 350 3.2. Optical Transmitters and Receivers 352 WDM optical systems make use of optical transmitters and receivers 353 utilizing different wavelengths (frequencies). Some transmitters are 354 manufactured for a specific wavelength of operation, that is, the 355 manufactured frequency cannot be changed. First introduced to reduce 356 inventory costs, tunable optical transmitters and receivers are 357 deployed in some systems, and allow flexibility in the wavelength 358 used for optical transmission/reception. Such tunable optics aid in 359 path selection. 361 Fundamental modeling parameters from the control plane perspective 362 optical transmitters and receivers are: 364 o Tunable: Do the transmitter and receivers operate at variable or 365 fixed wavelength. 367 o Tuning range: This is the frequency or wavelength range over which 368 the optics can be tuned. With the fixed mapping of labels to 369 lambdas as proposed in [Otani] this can be expressed as a tuple 370 (lambda1, lambda2) or (freq1, freq2) where lambda1 and lambda2 or 371 freq1 and freq2 are the labels representing the lower and upper 372 bounds in wavelength. 374 o Tuning time: Tuning times highly depend on the technology used. 375 Thermal drift based tuning may take seconds to stabilize, whilst 376 electronic tuning might provide sub-ms tuning times. Depending on 377 the application this might be critical. For example, thermal drift 378 might not be usable for fast protection applications. 380 o Spectral characteristics and stability: The spectral shape of a 381 laser's emissions and its frequency stability put limits on 382 various properties of the overall WDM system. One relatively easy 383 to characterize constraint is the closest channel spacing with 384 which the transmitter can be used. 386 Note that ITU-T recommendations specify many aspects of an optical 387 transmitter. Many of these parameters, such as spectral 388 characteristics and stability, are used in the design of WDM 389 subsystems consisting of transmitters, WDM links and receivers 390 however they do not furnish additional information that will 391 influence the Label Switched Path (LSP) provisioning in a properly 392 designed system. 394 Also note that optical components can degrade and fail over time. 395 This presents the possibility of the failure of a LSP (optical path) 396 without either a node or link failure. Hence, additional mechanisms 397 may be necessary to detect and differentiate this failure from the 398 others, e.g., one doesn't want to initiate mesh restoration if the 399 source transmitter has failed, since the optical transmitter will 400 still be failed on the alternate optical path. 402 3.3. Optical Signals in WSONs 404 In WSONs the fundamental unit of switching is intuitively that of a 405 "wavelength". The transmitters and receivers in these networks will 406 deal with one wavelength at a time, while the switching systems 407 themselves can deal with multiple wavelengths at a time. Hence 408 multichannel DWDM networks with single channel interfaces are the 409 prime focus of this document as opposed to multi-channel interfaces. 410 Interfaces of this type are defined in ITU-T recommendations 411 [G.698.1] and [G.698.2]. Key non-impairment related parameters 412 defined in [G.698.1] and [G.698.2] are: 414 (a) Minimum channel spacing (GHz) 415 (b) Minimum and maximum central frequency 417 (c) Bit-rate/Line coding (modulation) of optical tributary signals 419 For the purposes of modeling the WSON in the control plane, (a) and 420 (b) are considered as properties of the link and restrictions on the 421 GMPLS labels while (c) is a property of the "signal". 423 3.3.1. Optical Tributary Signals 425 The optical interface specifications [G.698.1], [G.698.2], and 426 [G.959.1] all use the concept of an optical tributary signal which is 427 defined as "a single channel signal that is placed within an optical 428 channel for transport across the optical network". Note the use of 429 the qualifier "tributary" to indicate that this is a single channel 430 entity and not a multichannel optical signal. 432 There are currently a number of different types of optical tributary 433 signals, which are known as "optical tributary signal classes". These 434 are currently characterized by a modulation format and bit rate range 435 [G.959.1]: 437 (a) Optical tributary signal class NRZ 1.25G 439 (b) Optical tributary signal class NRZ 2.5G 441 (c) Optical tributary signal class NRZ 10G 443 (d) Optical tributary signal class NRZ 40G 445 (e) Optical tributary signal class RZ 40G 447 Note that with advances in technology more optical tributary signal 448 classes may be added and that this is currently an active area for 449 development and standardization. In particular at the 40G rate there 450 are a number of non-standardized advanced modulation formats that 451 have seen significant deployment including Differential Phase Shift 452 Keying (DPSK) and Phase Shaped Binary Transmission (PSBT). 454 According to [G.698.2] it is important to fully specify the bit rate 455 of the optical tributary signal. Hence it is seen that modulation 456 format (optical tributary signal class) and bit rate are key 457 parameters in characterizing the optical tributary signal. 459 3.3.2. WSON Signal Characteristics 461 An optical tributary signal referenced in ITU-T [G.698.1] and 462 [G.698.2] is referred to as the "signal" in this document. This 463 corresponds to the "lambda" LSP in GMPLS. For signal compatibility 464 purposes with electro-optical network elements, the following signal 465 characteristics are considered: 467 1. Optical tributary signal class (modulation format). 468 2. FEC: whether forward error correction is used in the digital stream 469 and what type of error correcting code is used. 470 3. Center frequency (wavelength). 471 4. Bit rate. 472 5. G-PID: general protocol identifier for the information format. 474 The first three items on this list can change as a WSON signal 475 traverses the optical network with elements that include 476 regenerators, Optical-to-Electrical (OEO) switches, or wavelength 477 converters. 479 Bit rate and G-PID would not change since they describe the encoded 480 bit stream. A set of G-PID values is already defined for lambda 481 switching in [RFC3471] and [RFC4328]. 483 Note that a number of non-standard or proprietary modulation formats 484 and FEC codes are commonly used in WSONs. For some digital bit 485 streams the presence of Forward Error Correction (FEC) can be 486 detected, e.g., in [G.707] this is indicated in the signal itself via 487 the FEC Status Indication (FSI) byte, while in [G.709] this can be 488 inferred from whether the FEC field of the Optical Channel Transport 489 Unit-k (OTUk) is all zeros or not. 491 3.4. ROADMs, OXCs, Splitters, Combiners and FOADMs 493 Definitions of various optical devices such as ROADMs, Optical Cross- 494 connects (OXCs), splitters, combiners and Fixed Optical Add-Drop 495 Multiplexers (FOADMs) and their parameters can be found in [G.671]. 496 Only a subset of these relevant to the control plane and their non- 497 impairment related properties are considered in the following 498 sections. 500 3.4.1. Reconfigurable Add/Drop Multiplexers and OXCs 502 ROADMs are available in different forms and technologies. This is a 503 key technology that allows wavelength based optical switching. A 504 classic degree-2 ROADM is shown in Figure 1. 506 Line side input +---------------------+ Line side output 507 --->| |---> 508 | | 509 | ROADM | 510 | | 511 | | 512 +---------------------+ 513 | | | | o o o o 514 | | | | | | | | 515 O O O O | | | | 516 Tributary Side: Drop (output) Add (input) 518 Figure 1. Degree-2 unidirectional ROADM 520 The key feature across all ROADM types is their highly asymmetric 521 switching capability. In the ROADM of Figure 1, signals introduced 522 via the add ports can only be sent on the line side output port and 523 not on any of the drop ports. The term "degree" is used to refer to 524 the number of line side ports (input and output) of a ROADM, and does 525 not include the number of "add" or "drop" ports. The add and drop 526 ports are sometimes also called tributary ports. As the degree of the 527 ROADM increases beyond two it can have properties of both a switch 528 (OXC) and a multiplexer and hence it is necessary to know the 529 switched connectivity offered by such a network element to 530 effectively utilize it. A straightforward way to represent this is 531 via a "switched connectivity" matrix A where Amn = 0 or 1, depending 532 upon whether a wavelength on input port m can be connected to output 533 port n [Imajuku]. For the ROADM shown in Figure 1 the switched 534 connectivity matrix can be expressed as: 536 Input Output Port 537 Port #1 #2 #3 #4 #5 538 -------------- 539 #1: 1 1 1 1 1 540 #2 1 0 0 0 0 541 A = #3 1 0 0 0 0 542 #4 1 0 0 0 0 543 #5 1 0 0 0 0 545 Where input ports 2-5 are add ports, output ports 2-5 are drop ports 546 and input port #1 and output port #1 are the line side (WDM) ports. 548 For ROADMs, this matrix will be very sparse, and for OXCs the matrix 549 will be very dense. Compact encodings and examples, including high 550 degree ROADMs/OXCs, are given in [Gen-Encode]. A degree-4 ROADM is 551 shown in Figure 2. 553 +-----------------------+ 554 Line side-1 --->| |---> Line side-2 555 Input (I1) | | Output (E2) 556 Line side-1 <---| |<--- Line side-2 557 Output (E1) | | Input (I2) 558 | ROADM | 559 Line side-3 --->| |---> Line side-4 560 Input (I3) | | Output (E4) 561 Line side-3 <---| |<--- Line side-4 562 Output (E3) | | Input (I4) 563 | | 564 +-----------------------+ 565 | O | O | O | O 566 | | | | | | | | 567 O | O | O | O | 568 Tributary Side: E5 I5 E6 I6 E7 I7 E8 I8 570 Figure 2. Degree-4 bidirectional ROADM 572 Note that this example is 4-degree example with one (potentially 573 multi-channel) add/drop per line side port. 575 Note also that the connectivity constraints for typical ROADM designs 576 are "bidirectional", i.e. if input port X can be connected to output 577 port Y, typically input port Y can be connected to output port X, 578 assuming the numbering is done in such a way that input X and output 579 X correspond to the same line side direction or the same add/drop 580 port. This makes the connectivity matrix symmetrical as shown below. 582 Input Output Port 583 Port E1 E2 E3 E4 E5 E6 E7 E8 584 ----------------------- 585 I1 0 1 1 1 0 1 0 0 586 I2 1 0 1 1 0 0 1 0 587 A = I3 1 1 0 1 1 0 0 0 588 I4 1 1 1 0 0 0 0 1 589 I5 0 0 1 0 0 0 0 0 590 I6 1 0 0 0 0 0 0 0 591 I7 0 1 0 0 0 0 0 0 592 I8 0 0 0 1 0 0 0 0 594 Where I5/E5 are add/drop ports to/from line side-3, I6/E6 are 595 add/drop ports to/from line side-1, I7/E7 are add/drop ports to/from 596 line side-2 and I8/E8 are add/drop ports to/from line side-4. Note 597 that diagonal elements are zero since loopback is not supported in 598 the example. If ports support loopback, diagonal elements would be 599 set to one. 601 Additional constraints may also apply to the various ports in a 602 ROADM/OXC. The following restrictions and terms may be used: 604 Colored port: an input or more typically an output (drop) port 605 restricted to a single channel of fixed wavelength. 607 Colorless port: an input or more typically an output (drop) port 608 restricted to a single channel of arbitrary wavelength. 610 In general a port on a ROADM could have any of the following 611 wavelength restrictions: 613 o Multiple wavelengths, full range port. 615 o Single wavelength, full range port. 617 o Single wavelength, fixed lambda port. 619 o Multiple wavelengths, reduced range port (for example wave band 620 switching). 622 To model these restrictions it is necessary to have two pieces of 623 information for each port: (a) number of wavelengths, (b) wavelength 624 range and spacing. Note that this information is relatively static. 625 More complicated wavelength constraints are modeled in [WSON-Info]. 627 3.4.2. Splitters 629 An optical splitter consists of a single input port and two or more 630 output ports. The input optical signaled is essentially copied (with 631 power loss) to all output ports. 633 Using the modeling notions of Section 3.4.1. (Reconfigurable Add/Drop 634 Multiplexers and OXCs) the input and output ports of a splitter would 635 have the same wavelength restrictions. In addition a splitter is 636 modeled by a connectivity matrix Amn as follows: 638 Input Output Port 639 Port #1 #2 #3 ... #N 640 ----------------- 641 A = #1 1 1 1 ... 1 643 The difference from a simple ROADM is that this is not a switched 644 connectivity matrix but the fixed connectivity matrix of the device. 646 3.4.3. Combiners 648 An optical combiner is a device that combines the optical wavelengths 649 carried by multiple input ports into a single multi-wavelength output 650 port. The various ports may have different wavelength restrictions. 651 It is generally the responsibility of those using the combiner to 652 assure that wavelength collision does not occur on the output port. 653 The fixed connectivity matrix Amn for a combiner would look like: 655 Input Output Port 656 Port #1 657 --- 658 #1: 1 659 #2 1 660 A = #3 1 661 ... 1 662 #N 1 664 3.4.4. Fixed Optical Add/Drop Multiplexers 666 A fixed optical add/drop multiplexer can alter the course of an input 667 wavelength in a preset way. In particular a given wavelength (or 668 waveband) from a line side input port would be dropped to a fixed 669 "tributary" output port. Depending on the device's construction that 670 same wavelength may or may not also be sent out the line side output 671 port. This is commonly referred to as "drop and continue" operation. 672 There also may exist tributary input ports ("add" ports) whose 673 signals are combined with each other and other line side signals. 675 In general, to represent the routing properties of an FOADM it is 676 necessary to have both a fixed connectivity matrix Amn as previously 677 discussed and the precise wavelength restrictions for all input and 678 output ports. From the wavelength restrictions on the tributary 679 output ports, what wavelengths have been selected can be derived. 680 From the wavelength restrictions on the tributary input ports, it can 681 be seen which wavelengths have been added to the line side output 682 port. Finally from the added wavelength information and the line side 683 output wavelength restrictions it can be inferred which wavelengths 684 have been continued. 686 To summarize, the modeling methodology introduced in Section 3.4.1. 687 (Reconfigurable Add/Drop Multiplexers and OXCs) consisting of a 688 connectivity matrix and port wavelength restrictions can be used to 689 describe a large set of fixed optical devices such as combiners, 690 splitters and FOADMs. Hybrid devices consisting of both switched and 691 fixed parts are modeled in [WSON-Info]. 693 3.5. Electro-Optical Systems 695 This section describes how Electro-Optical Systems (e.g., OEO 696 switches, wavelength converters, and regenerators) interact with the 697 WSON signal characteristics listed in Section 3.3.2. (WSON Signal 698 Characteristics) OEO switches, wavelength converters and regenerators 699 all share a similar property: they can be more or less "transparent" 700 to an "optical signal" depending on their functionality and/or 701 implementation. Regenerators have been fairly well characterized in 702 this regard and hence their properties can be described first. 704 3.5.1. Regenerators 706 The various approaches to regeneration are discussed in ITU-T G.872 707 Annex A [G.872]. They map a number of functions into the so-called 708 1R, 2R and 3R categories of regenerators as summarized in Table 1 709 below: 711 Table 1. Regenerator functionality mapped to general regenerator 712 classes from [G.872]. 714 --------------------------------------------------------------------- 715 1R | Equal amplification of all frequencies within the amplification 716 | bandwidth. There is no restriction upon information formats. 717 +----------------------------------------------------------------- 718 | Amplification with different gain for frequencies within the 719 | amplification bandwidth. This could be applied to both single- 720 | channel and multi-channel systems. 721 +----------------------------------------------------------------- 722 | Dispersion compensation (phase distortion). This analogue 723 | process can be applied in either single-channel or multi- 724 | channel systems. 725 --------------------------------------------------------------------- 726 2R | Any or all 1R functions. Noise suppression. 727 +----------------------------------------------------------------- 728 | Digital reshaping (Schmitt Trigger function) with no clock 729 | recovery. This is applicable to individual channels and can be 730 | used for different bit rates but is not transparent to line 731 | coding (modulation). 732 -------------------------------------------------------------------- 733 3R | Any or all 1R and 2R functions. Complete regeneration of the 734 | pulse shape including clock recovery and retiming within 735 | required jitter limits. 736 -------------------------------------------------------------------- 738 From this table it is seen that 1R regenerators are generally 739 independent of signal modulation format (also known as line coding), 740 but may work over a limited range of wavelength/frequencies. 2R 741 regenerators are generally applicable to a single digital stream and 742 are dependent upon modulation format (line coding) and to a lesser 743 extent are limited to a range of bit rates (but not a specific bit 744 rate). Finally, 3R regenerators apply to a single channel, are 745 dependent upon the modulation format and generally sensitive to the 746 bit rate of digital signal, i.e., either are designed to only handle 747 a specific bit rate or need to be programmed to accept and regenerate 748 a specific bit rate. In all these types of regenerators the digital 749 bit stream contained within the optical or electrical signal is not 750 modified. 752 It is common for regenerators to modify the digital bit stream for 753 performance monitoring and fault management purposes. Synchronous 754 Optical Networking (SONET), Synchronous Digital Hierarchy (SDH) and 755 Interfaces for the Optical Transport Network (G.709) all have digital 756 signal "envelopes" designed to be used between "regenerators" (in 757 this case 3R regenerators). In SONET this is known as the "section" 758 signal, in SDH this is known as the "regenerator section" signal, in 759 G.709 this is known as an OTUk. These signals reserve a portion of 760 their frame structure (known as overhead) for use by regenerators. 761 The nature of this overhead is summarized in Table 2 below. 763 Table 2. SONET, SDH, and G.709 regenerator related overhead. 765 +-----------------------------------------------------------------+ 766 |Function | SONET/SDH | G.709 OTUk | 767 | | Regenerator | | 768 | | Section | | 769 |------------------+----------------------+-----------------------| 770 |Signal | J0 (section | Trail Trace | 771 |Identifier | trace) | Identifier (TTI) | 772 |------------------+----------------------+-----------------------| 773 |Performance | BIP-8 (B1) | BIP-8 (within SM) | 774 |Monitoring | | | 775 |------------------+----------------------+-----------------------| 776 |Management | D1-D3 bytes | GCC0 (general | 777 |Communications | | communications | 778 | | | channel) | 779 |------------------+----------------------+-----------------------| 780 |Fault Management | A1, A2 framing | FAS (frame alignment | 781 | | bytes | signal), BDI(backward| 782 | | | defect indication)BEI| 783 | | | (backward error | 784 | | | indication) | 785 +------------------+----------------------+-----------------------| 786 |Forward Error | P1,Q1 bytes | OTUk FEC | 787 |Correction (FEC) | | | 788 +-----------------------------------------------------------------+ 790 In the previous table it is seen that frame alignment, signal 791 identification, and FEC are supported. What table 2 also shows by its 792 omission is that no switching or multiplexing occurs at this layer. 793 This is a significant simplification for the control plane since 794 control plane standards require a multi-layer approach when there are 795 multiple switching layers, but not for "layering" to provide the 796 management functions of Table 2. That is, many existing technologies 797 covered by GMPLS contain extra management related layers that are 798 essentially ignored by the control plane (though not by the 799 management plane!). Hence, the approach here is to include 800 regenerators and other devices at the WSON layer unless they provide 801 higher layer switching and then a multi-layer or multi-region 802 approach [RFC5212] is called for. However, this can result in 803 regenerators having a dependence on the client signal type. 805 Hence depending upon the regenerator technology the following 806 constraints may be imposed by a regenerator device: 808 Table 3. Regenerator Compatibility Constraints. 810 +--------------------------------------------------------+ 811 | Constraints | 1R | 2R | 3R | 812 +--------------------------------------------------------+ 813 | Limited Wavelength Range | x | x | x | 814 +--------------------------------------------------------+ 815 | Modulation Type Restriction | | x | x | 816 +--------------------------------------------------------+ 817 | Bit Rate Range Restriction | | x | x | 818 +--------------------------------------------------------+ 819 | Exact Bit Rate Restriction | | | x | 820 +--------------------------------------------------------+ 821 | Client Signal Dependence | | | x | 822 +--------------------------------------------------------+ 824 Note that the limited wavelength range constraint can be modeled for 825 GMPLS signaling with the label set defined in [RFC3471] and that the 826 modulation type restriction constraint includes FEC. 828 3.5.2. OEO Switches 830 A common place where OEO processing may take place is within WSON 831 switches that utilize (or contain) regenerators. This may be to 832 convert the signal to an electronic form for switching then 833 reconverting to an optical signal prior to output from the switch. 834 Another common technique is to add regenerators to restore signal 835 quality either before or after optical processing (switching). In 836 the former case the regeneration is applied to adapt the signal to 837 the switch fabric regardless of whether or not it is needed from a 838 signal quality perspective. 840 In either case these optical switches have essentially the same 841 compatibility constraints as those which are described for 842 regenerators in Table 3. 844 3.6. Wavelength Converters 846 Wavelength converters take an input optical signal at one wavelength 847 and emit an equivalent content optical signal at another wavelength 848 on output. There are multiple approaches to building wavelength 849 converters. One approach is based on OEO conversion with fixed or 850 tunable optics on output. This approach can be dependent upon the 851 signal rate and format, i.e., this is basically an electrical 852 regenerator combined with a laser/receiver. Hence, this type of 853 wavelength converter has signal processing restrictions that are 854 essentially the same as those described for regenerators in Table 3 855 of section 3.5.1. 857 Another approach performs the wavelength conversion optically via 858 non-linear optical effects, similar in spirit to the familiar 859 frequency mixing used in radio frequency systems, but significantly 860 harder to implement. Such processes/effects may place limits on the 861 range of achievable conversion. These may depend on the wavelength of 862 the input signal and the properties of the converter as opposed to 863 only the properties of the converter in the OEO case. Different WSON 864 system designs may choose to utilize this component to varying 865 degrees or not at all. 867 Current or envisioned contexts for wavelength converters are: 869 1. Wavelength conversion associated with OEO switches and fixed or 870 tunable optics. In this case there are typically multiple 871 converters available since each use of an OEO switch can be thought 872 of as a potential wavelength converter. 874 2. Wavelength conversion associated with ROADMs/OXCs. In this case 875 there may be a limited pool of wavelength converters available. 876 Conversion could be either all optical or via an OEO method. 878 3. Wavelength conversion associated with fixed devices such as FOADMs. 879 In this case there may be a limited amount of conversion. Also in 880 this case the conversion may be used as part of optical path 881 routing. 883 Based on the above considerations, wavelength converters are modeled 884 as follows: 886 1. Wavelength converters can always be modeled as associated with 887 network elements. This includes fixed wavelength routing elements. 889 2. A network element may have full wavelength conversion capability, 890 i.e., any input port and wavelength, or a limited number of 891 wavelengths and ports. On a box with a limited number of 892 converters there also may exist restrictions on which ports can 893 reach the converters. Hence regardless of where the converters 894 actually are they can be associated with input ports. 896 3. Wavelength converters have range restrictions that are either 897 independent or dependent upon the input wavelength. 899 In WSONs where wavelength converters are sparse an optical path may 900 appear to loop or "backtrack" upon itself in order to reach a 901 wavelength converter prior to continuing on to its destination. The 902 lambda used on input to the wavelength converter would be different 903 from the lambda coming back from the wavelength converter. 905 A model for an individual O-E-O wavelength converter would consist 906 of: 908 o Input lambda or frequency range. 910 o Output lambda or frequency range. 912 3.6.1. Wavelength Converter Pool Modeling 914 A WSON node may include multiple wavelength converters. These are 915 usually arranged into some type of pool to promote resource sharing. 916 There are a number of different approaches used in the design of 917 switches with converter pools. However, from the point of view of 918 path computation it is necessary to know the following: 920 1. The nodes that support wavelength conversion. 922 2. The accessibility and availability of a wavelength converter to 923 convert from a given input wavelength on a particular input port 924 to a desired output wavelength on a particular output port. 926 3. Limitations on the types of signals that can be converted and the 927 conversions that can be performed. 929 To model point 2 above, a technique similar to that used to model 930 ROADMs and optical switches can be used, i.e., matrices to indicate 931 possible connectivity along with wavelength constraints for 932 links/ports. Since wavelength converters are considered a scarce 933 resource it will be desirable to include as a minimum the usage state 934 of individual wavelength converters in the pool. 936 A three stage model is used as shown schematically in Figure 3. 937 (Schematic diagram of wavelength converter pool model). This model 938 represents N input ports (fibers), P wavelength converters, and M 939 output ports (fibers). Since not all input ports can necessarily 940 reach the converter pool, the model starts with a wavelength pool 941 input matrix WI(i,p) = {0,1} where input port i can potentially reach 942 wavelength converter p. 944 Since not all wavelengths can necessarily reach all the converters or 945 the converters may have limited input wavelength range there is a set 946 of input port constraints for each wavelength converter. Currently it 947 is assumed that a wavelength converter can only take a single 948 wavelength on input. Each wavelength converter input port constraint 949 can be modeled via a wavelength set mechanism. 951 Next a state vector WC(j) = {0,1} dependent upon whether wavelength 952 converter j in the pool is in use. This is the only state kept in the 953 converter pool model. This state is not necessary for modeling 954 "fixed" transponder system, i.e., systems where there is no sharing. 955 In addition, this state information may be encoded in a much more 956 compact form depending on the overall connectivity structure [Gen- 957 Encode]. 959 After that, a set of wavelength converter output wavelength 960 constraints is used. These constraints indicate what wavelengths a 961 particular wavelength converter can generate or are restricted to 962 generating due to internal switch structure. 964 Finally, a wavelength pool output matrix WE(p,k) = {0,1} indicating 965 whether the output from wavelength converter p can reach output port 966 k. Examples of this method being used to model wavelength converter 967 pools for several switch architectures are given in reference [Gen- 968 Encode]. 970 I1 +-------------+ +-------------+ E1 971 ----->| | +--------+ | |-----> 972 I2 | +------+ WC #1 +-------+ | E2 973 ----->| | +--------+ | |-----> 974 | Wavelength | | Wavelength | 975 | Converter | +--------+ | Converter | 976 | Pool +------+ WC #2 +-------+ Pool | 977 | | +--------+ | | 978 | Input | | Output | 979 | Connection | . | Connection | 980 | Matrix | . | Matrix | 981 | | . | | 982 | | | | 983 IN | | +--------+ | | EM 984 ----->| +------+ WC #P +-------+ |-----> 985 | | +--------+ | | 986 +-------------+ ^ ^ +-------------+ 987 | | 988 | | 989 | | 990 | | 992 Input wavelength Output wavelength 993 constraints for constraints for 994 each converter each converter 996 Figure 3. Schematic diagram of wavelength converter pool model. 998 Figure 4 below shows a simple optical switch in a four wavelength 999 DWDM system sharing wavelength converters in a general shared "per 1000 node" fashion. 1002 +-----------+ ___________ +------+ 1003 | |--------------------------->| | 1004 | |--------------------------->| C | 1005 /| | |--------------------------->| o | E1 1006 I1 /D+--->| |--------------------------->| m | 1007 + e+--->| | | b |====> 1008 ====>| M| | Optical | +-----------+ +----+ | i | 1009 + u+--->| Switch | | WC Pool | |O S|-->| n | 1010 \x+--->| | | +-----+ | |p w|-->| e | 1011 \| | +----+->|WC #1|--+->|t i| | r | 1012 | | | +-----+ | |i t| +------+ 1013 | | | | |c c| +------+ 1014 /| | | | +-----+ | |a h|-->| | 1015 I2 /D+--->| +----+->|WC #2|--+->|l |-->| C | E2 1016 + e+--->| | | +-----+ | | | | o | 1017 ====>| M| | | +-----------+ +----+ | m |====> 1018 + u+--->| | | b | 1019 \x+--->| |--------------------------->| i | 1020 \| | |--------------------------->| n | 1021 | |--------------------------->| e | 1022 |___________|--------------------------->| r | 1023 +-----------+ +------+ 1025 Figure 4. An optical switch featuring a shared per node wavelength 1026 converter pool architecture. 1028 In this case the input and output pool matrices are simply: 1030 +-----+ +-----+ 1031 | 1 1 | | 1 1 | 1032 WI =| |, WE =| | 1033 | 1 1 | | 1 1 | 1034 +-----+ +-----+ 1036 Figure 5 shows a different wavelength pool architecture known as 1037 "shared per fiber". In this case the input and output pool matrices 1038 are simply: 1040 +-----+ +-----+ 1041 | 1 1 | | 1 0 | 1042 WI =| |, WE =| | 1043 | 1 1 | | 0 1 | 1044 +-----+ +-----+ 1045 +-----------+ +------+ 1046 | |--------------------------->| | 1047 | |--------------------------->| C | 1048 /| | |--------------------------->| o | E1 1049 I1 /D+--->| |--------------------------->| m | 1050 + e+--->| | | b |====> 1051 ====>| M| | Optical | +-----------+ | i | 1052 + u+--->| Switch | | WC Pool | | n | 1053 \x+--->| | | +-----+ | | e | 1054 \| | +----+->|WC #1|--+---------->| r | 1055 | | | +-----+ | +------+ 1056 | | | | +------+ 1057 /| | | | +-----+ | | | 1058 I2 /D+--->| +----+->|WC #2|--+---------->| C | E2 1059 + e+--->| | | +-----+ | | o | 1060 ====>| M| | | +-----------+ | m |====> 1061 + u+--->| | | b | 1062 \x+--->| |--------------------------->| i | 1063 \| | |--------------------------->| n | 1064 | |--------------------------->| e | 1065 |___________|--------------------------->| r | 1066 +-----------+ +------+ 1067 Figure 5. An optical switch featuring a shared per fiber wavelength 1068 converter pool architecture. 1070 3.7. Characterizing Electro-Optical Network Elements 1072 In this section electro-optical WSON network elements are 1073 characterized by the three key functional components: input 1074 constraints, output constraints and processing capabilities. 1076 WSON Network Element 1077 +-----------------------+ 1078 WSON Signal | | | | WSON Signal 1079 | | | | 1080 ---------------> | | | | -----------------> 1081 | | | | 1082 +-----------------------+ 1083 <-----> <-------> <-----> 1085 Input Processing Output 1087 Figure 6. WSON Network Element 1088 3.7.1. Input Constraints 1090 Section 3. (Wavelength Switched Optical Networks) discussed the basic 1091 properties regenerators, OEO switches and wavelength converters. From 1092 these the following possible types of input constraints and 1093 properties are derived: 1095 1. Acceptable Modulation formats. 1097 2. Client Signal (G-PID) restrictions. 1099 3. Bit Rate restrictions. 1101 4. FEC coding restrictions. 1103 5. Configurability: (a) none, (b) self-configuring, (c) required. 1105 These constraints are represented via simple lists. Note that the 1106 device may need to be "provisioned" via signaling or some other means 1107 to accept signals with some attributes versus others. In other cases 1108 the devices maybe relatively transparent to some attributes, e.g., 1109 such as a 2R regenerator to bit rate. Finally, some devices may be 1110 able to auto-detect some attributes and configure themselves, e.g., a 1111 3R regenerator with bit rate detection mechanisms and flexible phase 1112 locking circuitry. To account for these different cases item 5 has 1113 been added, which describes the devices configurability. 1115 Note that such input constraints also apply to the termination of the 1116 WSON signal. 1118 3.7.2. Output Constraints 1120 None of the network elements considered here modifies either the bit 1121 rate or the basic type of the client signal. However, they may modify 1122 the modulation format or the FEC code. Typically the following types 1123 of output constraints are seen: 1125 1. Output modulation is the same as input modulation (default). 1127 2. A limited set of output modulations is available. 1129 3. Output FEC is the same as input FEC code (default). 1131 4. A limited set of output FEC codes is available. 1133 Note that in cases (2) and (4) above, where there is more than one 1134 choice in the output modulation or FEC code then the network element 1135 will need to be configured on a per LSP basis as to which choice to 1136 use. 1138 3.7.3. Processing Capabilities 1140 A general WSON network element (NE) can perform a number of signal 1141 processing functions including: 1143 (A) Regeneration (possibly different types). 1145 (B) Fault and Performance Monitoring. 1147 (C) Wavelength Conversion. 1149 (D) Switching. 1151 An NE may or may not have the ability to perform regeneration (of the 1152 one of the types previously discussed). In addition some nodes may 1153 have limited regeneration capability, i.e., a shared pool, which may 1154 be applied to selected signals traversing the NE. Hence to describe 1155 the regeneration capability of a link or node it is necessary to have 1156 at a minimum: 1158 1. Regeneration capability: (a)fixed, (b) selective, (c) none. 1160 2. Regeneration type: 1R, 2R, 3R. 1162 3. Regeneration pool properties for the case of selective 1163 regeneration (input and output restrictions, availability). 1165 Note that the properties of shared regenerator pools would be 1166 essentially the same as that of wavelength converter pools modeled in 1167 section 3.6.1. (Wavelength Pool Convertor Modeling). 1169 Item (B), fault and performance monitoring, is typically outside the 1170 scope of the control plane. However, when the operations are to be 1171 performed on an LSP basis or on part of an LSP then the control plane 1172 can be of assistance in their configuration. Per LSP, per node, fault 1173 and performance monitoring examples include setting up a "section 1174 trace" (a regenerator overhead identifier) between two nodes, or 1175 intermediate optical performance monitoring at selected nodes along a 1176 path. 1178 4. Routing and Wavelength Assignment and the Control Plane 1180 From a control plane perspective, a wavelength-convertible network 1181 with full wavelength-conversion capability at each node can be 1182 controlled much like a packet MPLS-labeled network or a circuit- 1183 switched Time-division multiplexing (TDM) network with full time slot 1184 interchange capability is controlled. In this case, the path 1185 selection process needs to identify the Traffic Engineered (TE) links 1186 to be used by an optical path, and wavelength assignment can be made 1187 on a hop-by-hop basis. 1189 However, in the case of an optical network without wavelength 1190 converters, an optical path needs to be routed from source to 1191 destination and must use a single wavelength that is available along 1192 that path without "colliding" with a wavelength used by any other 1193 optical path that may share an optical fiber. This is sometimes 1194 referred to as a "wavelength continuity constraint". 1196 In the general case of limited or no wavelength converters the 1197 computation of both the links and wavelengths is known as RWA. 1199 The inputs to basic RWA are the requested optical path's source and 1200 destination, the network topology, the locations and capabilities of 1201 any wavelength converters, and the wavelengths available on each 1202 optical link. The output from an algorithm providing RWA is an 1203 explicit route through ROADMs, a wavelength for optical transmitter, 1204 and a set of locations (generally associated with ROADMs or switches) 1205 where wavelength conversion is to occur and the new wavelength to be 1206 used on each component link after that point in the route. 1208 It is to be noted that the choice of specific RWA algorithm is out of 1209 the scope for this document. However there are a number of different 1210 approaches to dealing with RWA algorithm that can affect the division 1211 of effort between path computation/routing and signaling. 1213 4.1. Architectural Approaches to RWA 1215 Two general computational approaches are taken to performing RWA. 1216 Some algorithms utilize a two-step procedure of path selection 1217 followed by wavelength assignment, and others perform RWA in a 1218 combined fashion. 1220 In the following, three different ways of performing RWA in 1221 conjunction with the control plane are considered. The choice of one 1222 of these architectural approaches over another generally impacts the 1223 demands placed on the various control plane protocols. The approaches 1224 are provided for reference purposes only, and other approaches are 1225 possible. 1227 4.1.1. Combined RWA (R&WA) 1229 In this case, a unique entity is in charge of performing routing and 1230 wavelength assignment. This approach relies on a sufficient knowledge 1231 of network topology, of available network resources and of network 1232 nodes' capabilities. This solution is compatible with most known RWA 1233 algorithms, and in particular those concerned with network 1234 optimization. On the other hand, this solution requires up-to-date 1235 and detailed network information. 1237 Such a computational entity could reside in two different places: 1239 o In a PCE which maintains a complete and updated view of network 1240 state and provides path computation services to nodes (PCCs). 1242 o In an ingress node, in which case all nodes have the R&WA 1243 functionality and network state is obtained by a periodic flooding 1244 of information provided by the other nodes. 1246 4.1.2. Separated R and WA (R+WA) 1248 In this case, one entity performs routing, while a second performs 1249 wavelength assignment. The first entity furnishes one or more paths 1250 to the second entity which will perform wavelength assignment and 1251 final path selection. 1253 As the entities computing the path and the wavelength assignment are 1254 separated, this constrains the class of RWA algorithms that may be 1255 implemented. Although it may seem that algorithms optimizing a joint 1256 usage of the physical and wavelength paths are excluded from this 1257 solution, many practical optimization algorithms only consider a 1258 limited set of possible paths, e.g., as computed via a k-shortest 1259 path algorithm. Hence, while there is no guarantee that the selected 1260 final route and wavelength offers the optimal solution, by allowing 1261 multiple routes to pass to the wavelength selection process 1262 reasonable optimization can be performed. 1264 The entity performing the routing assignment needs the topology 1265 information of the network, whereas the entity performing the 1266 wavelength assignment needs information on the network's available 1267 resources and specific network node capabilities. 1269 4.1.3. Routing and Distributed WA (R+DWA) 1271 In this case, one entity performs routing, while wavelength 1272 assignment is performed on a hop-by-hop, distributed manner along the 1273 previously computed path. This mechanism relies on updating of a list 1274 of potential wavelengths used to ensure conformance with the 1275 wavelength continuity constraint. 1277 As currently specified, the GMPLS protocol suite signaling protocol 1278 can accommodate such an approach. GMPLS, per [RFC3471], includes 1279 support for the communication of the set of labels (wavelengths) that 1280 may be used between nodes via a Label Set. When conversion is not 1281 performed at an intermediate node, a hop generates the Label Set it 1282 sends to the next hop based on the intersection of the Label Set 1283 received from the previous hop and the wavelengths available on the 1284 node's switch and ongoing interface. The generation of the outgoing 1285 Label Set is up to the node local policy (even if one expects a 1286 consistent policy configuration throughout a given transparency 1287 domain). When wavelength conversion is performed at an intermediate 1288 node, a new Label Set is generated. The egress node selects one label 1289 in the Label Set which it received; additionally the node can apply 1290 local policy during label selection. GMPLS also provides support for 1291 the signaling of bidirectional optical paths. 1293 Depending on these policies a wavelength assignment may not be found 1294 or one may be found that consumes too many conversion resources 1295 relative to what a dedicated wavelength assignment policy would have 1296 achieved. Hence, this approach may generate higher blocking 1297 probabilities in a heavily loaded network. 1299 This solution may be facilitated via signaling extensions which ease 1300 its functioning and possibly enhance its performance with respect to 1301 blocking probability. Note that this approach requires less 1302 information dissemination than the other techniques described. 1304 The first entity may be a PCE or the ingress node of the LSP. 1306 4.2. Conveying information needed by RWA 1308 The previous sections have characterized WSONs and optical path 1309 requests. In particular, high level models of the information used by 1310 RWA process were presented. This information can be viewed as either 1311 relatively static, i.e., changing with hardware changes (including 1312 possibly failures), or relatively dynamic, i.e., those that can 1313 change with optical path provisioning. The time requirement in which 1314 an entity involved in RWA process needs to be notified of such 1315 changes is fairly situational. For example, for network restoration 1316 purposes, learning of a hardware failure or of new hardware coming 1317 online to provide restoration capability can be critical. 1319 Currently there are various methods for communicating RWA relevant 1320 information, these include, but are not limited to: 1322 o Existing control plane protocols, i.e., GMPLS routing and 1323 signaling. Note that routing protocols can be used to convey both 1324 static and dynamic information. 1326 o Management protocols such as NetConf, SNMPv3, CLI, and CORBA. 1328 o Directory services and accompanying protocols. These are typically 1329 used for the dissemination of relatively static information. 1330 Directory services are not suited to manage information in dynamic 1331 and fluid environments. 1333 o Other techniques for dynamic information, e.g., sending 1334 information directly from NEs to PCE to avoid flooding. This would 1335 be useful if the number of PCEs is significantly less than number 1336 of WSON NEs. There may be other ways to limit flooding to 1337 "interested" NEs. 1339 Possible mechanisms to improve scaling of dynamic information 1340 include: 1342 o Tailor message content to WSON. For example the use of wavelength 1343 ranges, or wavelength occupation bit maps. 1345 o Utilize incremental updates if feasible. 1347 5. Modeling Examples and Control Plane Use Cases 1349 This section provides examples of the fixed and switched optical node 1350 and wavelength constraint models of Section 3. and use cases for WSON 1351 control plane path computation, establishment, rerouting, and 1352 optimization. 1354 5.1. Network Modeling for GMPLS/PCE Control 1356 Consider a network containing three routers (R1 through R3), eight 1357 WSON nodes (N1 through N8) and 18 links (L1 through L18) and one OEO 1358 converter (O1) in a topology shown below. 1360 +--+ +--+ +--+ +--------+ 1361 +-L3-+N2+-L5-+ +--------L12--+N6+--L15--+ N8 + 1362 | +--+ |N4+-L8---+ +--+ ++--+---++ 1363 | | +-L9--+| | | | 1364 +--+ +-+-+ ++-+ || | L17 L18 1365 | ++-L1--+ | | ++++ +----L16---+ | | 1366 |R1| | N1| L7 |R2| | | | 1367 | ++-L2--+ | | ++-+ | ++---++ 1368 +--+ +-+-+ | | | + R3 | 1369 | +--+ ++-+ | | +-----+ 1370 +-L4-+N3+-L6-+N5+-L10-+ ++----+ 1371 +--+ | +--------L11--+ N7 + 1372 +--+ ++---++ 1373 | | 1374 L13 L14 1375 | | 1376 ++-+ | 1377 |O1+-+ 1378 +--+ 1380 Figure 7. Routers and WSON nodes in a GMPLS and PCE Environment. 1382 5.1.1. Describing the WSON nodes 1384 The eight WSON nodes described in Figure 7 have the following 1385 properties: 1387 o Nodes N1, N2, N3 have FOADMs installed and can therefore only 1388 access a static and pre-defined set of wavelengths. 1390 o All other nodes contain ROADMs and can therefore access all 1391 wavelengths. 1393 o Nodes N4, N5, N7 and N8 are multi-degree nodes, allowing any 1394 wavelength to be optically switched between any of the links. Note 1395 however, that this does not automatically apply to wavelengths 1396 that are being added or dropped at the particular node. 1398 o Node N4 is an exception to that: This node can switch any 1399 wavelength from its add/drop ports to any of its output links (L5, 1400 L7 and L12 in this case). 1402 o The links from the routers are only able to carry one wavelength 1403 with the exception of links L8 and L9 which are capable to 1404 add/drop any wavelength. 1406 o Node N7 contains an OEO transponder (O1) connected to the node via 1407 links L13 and L14. That transponder operates in 3R mode and does 1408 not change the wavelength of the signal. Assume that it can 1409 regenerate any of the client signals, however only for a specific 1410 wavelength. 1412 Given the above restrictions, the node information for the eight 1413 nodes can be expressed as follows: (where ID == identifier, SCM == 1414 switched connectivity matrix, and FCM == fixed connectivity matrix). 1416 +ID+SCM +FCM + 1417 | | |L1 |L2 |L3 |L4 | | |L1 |L2 |L3 |L4 | | 1418 | |L1 |0 |0 |0 |0 | |L1 |0 |0 |1 |0 | | 1419 |N1|L2 |0 |0 |0 |0 | |L2 |0 |0 |0 |1 | | 1420 | |L3 |0 |0 |0 |0 | |L3 |1 |0 |0 |1 | | 1421 | |L4 |0 |0 |0 |0 | |L4 |0 |1 |1 |0 | | 1422 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1423 | | |L3 |L5 | | | | |L3 |L5 | | | | 1424 |N2|L3 |0 |0 | | | |L3 |0 |1 | | | | 1425 | |L5 |0 |0 | | | |L5 |1 |0 | | | | 1426 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1427 | | |L4 |L6 | | | | |L4 |L6 | | | | 1428 |N3|L4 |0 |0 | | | |L4 |0 |1 | | | | 1429 | |L6 |0 |0 | | | |L6 |1 |0 | | | | 1430 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1431 | | |L5 |L7 |L8 |L9 |L12| |L5 |L7 |L8 |L9 |L12| 1432 | |L5 |0 |1 |1 |1 |1 |L5 |0 |0 |0 |0 |0 | 1433 |N4|L7 |1 |0 |1 |1 |1 |L7 |0 |0 |0 |0 |0 | 1434 | |L8 |1 |1 |0 |1 |1 |L8 |0 |0 |0 |0 |0 | 1435 | |L9 |1 |1 |1 |0 |1 |L9 |0 |0 |0 |0 |0 | 1436 | |L12|1 |1 |1 |1 |0 |L12|0 |0 |0 |0 |0 | 1437 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1438 | | |L6 |L7 |L10|L11| | |L6 |L7 |L10|L11| | 1439 | |L6 |0 |1 |0 |1 | |L6 |0 |0 |1 |0 | | 1440 |N5|L7 |1 |0 |0 |1 | |L7 |0 |0 |0 |0 | | 1441 | |L10|0 |0 |0 |0 | |L10|1 |0 |0 |0 | | 1442 | |L11|1 |1 |0 |0 | |L11|0 |0 |0 |0 | | 1443 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1444 | | |L12|L15| | | | |L12|L15| | | | 1445 |N6|L12|0 |1 | | | |L12|0 |0 | | | | 1446 | |L15|1 |0 | | | |L15|0 |0 | | | | 1447 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1448 | | |L11|L13|L14|L16| | |L11|L13|L14|L16| | 1449 | |L11|0 |1 |0 |1 | |L11|0 |0 |0 |0 | | 1450 |N7|L13|1 |0 |0 |0 | |L13|0 |0 |1 |0 | | 1451 | |L14|0 |0 |0 |1 | |L14|0 |1 |0 |0 | | 1452 | |L16|1 |0 |1 |0 | |L16|0 |0 |1 |0 | | 1453 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1454 | | |L15|L16|L17|L18| | |L15|L16|L17|L18| | 1455 | |L15|0 |1 |0 |0 | |L15|0 |0 |0 |1 | | 1456 |N8|L16|1 |0 |0 |0 | |L16|0 |0 |1 |0 | | 1457 | |L17|0 |0 |0 |0 | |L17|0 |1 |0 |0 | | 1458 | |L18|0 |0 |0 |0 | |L18|1 |0 |1 |0 | | 1459 +--+---+---+---+---+---+---+---+---+---+---+---+---+ 1461 5.1.2. Describing the links 1463 For the following discussion some simplifying assumptions are made: 1465 o It is assumed that the WSON node support a total of four 1466 wavelengths designated WL1 through WL4. 1468 o It is assumed that the impairment feasibility of a path or path 1469 segment is independent from the wavelength chosen. 1471 For the discussion of RWA operation to build LSPs between two 1472 routers, the wavelength constraints on the links between the routers 1473 and the WSON nodes as well as the connectivity matrix of these links 1474 needs to be specified: 1476 +Link+WLs supported +Possible output links+ 1477 | L1 | WL1 | L3 | 1478 +----+-----------------+---------------------+ 1479 | L2 | WL2 | L4 | 1480 +----+-----------------+---------------------+ 1481 | L8 | WL1 WL2 WL3 WL4 | L5 L7 L12 | 1482 +----+-----------------+---------------------+ 1483 | L9 | WL1 WL2 WL3 WL4 | L5 L7 L12 | 1484 +----+-----------------+---------------------+ 1485 | L10| WL2 | L6 | 1486 +----+-----------------+---------------------+ 1487 | L13| WL1 WL2 WL3 WL4 | L11 L14 | 1488 +----+-----------------+---------------------+ 1489 | L14| WL1 WL2 WL3 WL4 | L13 L16 | 1490 +----+-----------------+---------------------+ 1491 | L17| WL2 | L16 | 1492 +----+-----------------+---------------------+ 1493 | L18| WL1 | L15 | 1494 +----+-----------------+---------------------+ 1496 Note that the possible output links for the links connecting to the 1497 routers is inferred from the switched connectivity matrix and the 1498 fixed connectivity matrix of the Nodes N1 through N8 and is show here 1499 for convenience, i.e., this information does not need to be repeated. 1501 5.2. RWA Path Computation and Establishment 1503 The calculation of optical impairment feasible routes is outside the 1504 scope of this document. In general optical impairment feasible routes 1505 serve as an input to RWA algorithm. 1507 For the example use case shown here, assume the following feasible 1508 routes: 1510 +Endpoint 1+Endpoint 2+Feasible Route + 1511 | R1 | R2 | L1 L3 L5 L8 | 1512 | R1 | R2 | L1 L3 L5 L9 | 1513 | R1 | R2 | L2 L4 L6 L7 L8 | 1514 | R1 | R2 | L2 L4 L6 L7 L9 | 1515 | R1 | R2 | L2 L4 L6 L10 | 1516 | R1 | R3 | L1 L3 L5 L12 L15 L18 | 1517 | R1 | N7 | L2 L4 L6 L11 | 1518 | N7 | R3 | L16 L17 | 1519 | N7 | R2 | L16 L15 L12 L9 | 1520 | R2 | R3 | L8 L12 L15 L18 | 1521 | R2 | R3 | L8 L7 L11 L16 L17 | 1522 | R2 | R3 | L9 L12 L15 L18 | 1523 | R2 | R3 | L9 L7 L11 L16 L17 | 1525 Given a request to establish a LSP between R1 and R2 RWA algorithm 1526 finds the following possible solutions: 1528 +WL + Path + 1529 | WL1| L1 L3 L5 L8 | 1530 | WL1| L1 L3 L5 L9 | 1531 | WL2| L2 L4 L6 L7 L8| 1532 | WL2| L2 L4 L6 L7 L9| 1533 | WL2| L2 L4 L6 L10 | 1535 Assume now that RWA algorithm yields WL1 and the Path L1 L3 L5 L8 for 1536 the requested LSP. 1538 Next, another LSP is signaled from R1 to R2. Given the established 1539 LSP using WL1, the following table shows the available paths: 1541 +WL + Path + 1542 | WL2| L2 L4 L6 L7 L9| 1543 | WL2| L2 L4 L6 L10 | 1545 Assume now that RWA algorithm yields WL2 and the path L2 L4 L6 L7 L9 1546 for the establishment of the new LSP. 1548 A LSP request -this time from R2 to R3 - can not be fulfilled since 1549 the only four possible paths (starting at L8 and L9) are already in 1550 use. 1552 5.3. Resource Optimization 1554 The preceding example gives rise to another use case: the 1555 optimization of network resources. Optimization can be achieved on a 1556 number of layers (e.g. through electrical or optical multiplexing of 1557 client signals) or by re-optimizing the solutions found by a RWA 1558 algorithm. 1560 Given the above example again, assume that a RWA algorithm should 1561 identify a path between R2 and R3. The only possible path to reach R3 1562 from R2 needs to use L9. L9 however is blocked by one of the LSPs 1563 from R1. 1565 5.4. Support for Rerouting 1567 It is also envisioned that the extensions to GMPLS and PCE support 1568 rerouting of wavelengths in case of failures. 1570 Assume for this discussion that the only two LSPs in use in the 1571 system are: 1573 LSP1: WL1 L1 L3 L5 L8 1575 LSP2: WL2 L2 L4 L6 L7 L9 1577 Assume furthermore that the link L5 fails. An RWA algorithm can now 1578 compute the following alternate path and establish that path: 1580 R1 -> N7 -> R2 1582 Level 3 regeneration will take place at N7, so that the complete path 1583 looks like this: 1585 R1 -> L2 L4 L6 L11 L13 -> O1 -> L14 L16 L15 L12 L9 -> R2 1587 5.5. Electro-Optical Networking Scenarios 1589 In the following various networking scenarios are considered 1590 involving regenerators, OEO switches and wavelength converters. These 1591 scenarios can be grouped roughly by type and number of extensions to 1592 the GMPLS control plane that would be required. 1594 5.5.1. Fixed Regeneration Points 1596 In the simplest networking scenario involving regenerators, 1597 regeneration is associated with a WDM link or an entire node and is 1598 not optional, i.e., all signals traversing the link or node will be 1599 regenerated. This includes OEO switches since they provide 1600 regeneration on every port. 1602 There may be input constraints and output constraints on the 1603 regenerators. Hence the path selection process will need to know from 1604 routing or other means the regenerator constraints so that it can 1605 choose a compatible path. For impairment aware routing and wavelength 1606 assignment (IA-RWA) the path selection process will also need to know 1607 which links/nodes provide regeneration. Even for "regular" RWA, this 1608 regeneration information is useful since wavelength converters 1609 typically perform regeneration and the wavelength continuity 1610 constraint can be relaxed at such a point. 1612 Signaling does not need to be enhanced to include this scenario since 1613 there are no reconfigurable regenerator options on input, output or 1614 with respect to processing. 1616 5.5.2. Shared Regeneration Pools 1618 In this scenario there are nodes with shared regenerator pools within 1619 the network in addition to fixed regenerators of the previous 1620 scenario. These regenerators are shared within a node and their 1621 application to a signal is optional. There are no reconfigurable 1622 options on either input or output. The only processing option is to 1623 "regenerate" a particular signal or not. 1625 Regenerator information in this case is used in path computation to 1626 select a path that ensures signal compatibility and IA-RWA criteria. 1628 To setup an LSP that utilizes a regenerator from a node with a shared 1629 regenerator pool it is necessary to indicate that regeneration is to 1630 take place at that particular node along the signal path. Such a 1631 capability currently does not exist in GMPLS signaling. 1633 5.5.3. Reconfigurable Regenerators 1635 This scenario is concerned with regenerators that require 1636 configuration prior to use on an optical signal. As discussed 1637 previously, this could be due to a regenerator that must be 1638 configured to accept signals with different characteristics, for 1639 regenerators with a selection of output attributes, or for 1640 regenerators with additional optional processing capabilities. 1642 As in the previous scenarios it is necessary to have information 1643 concerning regenerator properties for selection of compatible paths 1644 and for IA-RWA computations. In addition during LSP setup it is 1645 necessary to be able configure regenerator options at a particular 1646 node along the path. Such a capability currently does not exist in 1647 GMPLS signaling. 1649 5.5.4. Relation to Translucent Networks 1651 Networks that contain both transparent network elements such as 1652 reconfigurable optical add drop multiplexers (ROADMs) and electro- 1653 optical network elements such regenerators or OEO switches are 1654 frequently referred to as translucent optical networks. 1656 Three main types of translucent optical networks have been discussed: 1658 1. Transparent "islands" surrounded by regenerators. This is 1659 frequently seen when transitioning from a metro optical sub- 1660 network to a long haul optical sub-network. 1662 2. Mostly transparent networks with a limited number of OEO 1663 ("opaque") nodes strategically placed. This takes advantage of the 1664 inherent regeneration capabilities of OEO switches. In the 1665 planning of such networks one has to determine the optimal 1666 placement of the OEO switches. 1668 3. Mostly transparent networks with a limited number of optical 1669 switching nodes with "shared regenerator pools" that can be 1670 optionally applied to signals passing through these switches. 1671 These switches are sometimes called translucent nodes. 1673 All three types of translucent networks fit within the networking 1674 scenarios of Section 5.5.1. and Section 5.5.2. above. And hence, 1675 can be accommodated by the GMPLS extensions envisioned in this 1676 document. 1678 6. GMPLS and PCE Implications 1680 The presence and amount of wavelength conversion available at a 1681 wavelength switching interface has an impact on the information that 1682 needs to be transferred by the control plane (GMPLS) and the PCE 1683 architecture. Current GMPLS and PCE standards can address the full 1684 wavelength conversion case so the following will only address the 1685 limited and no wavelength conversion cases. 1687 6.1. Implications for GMPLS signaling 1689 Basic support for WSON signaling already exists in GMPLS with the 1690 lambda (value 9) LSP encoding type [RFC3471], or for G.709 compatible 1691 optical channels, the LSP encoding type (value = 13) "G.709 Optical 1692 Channel" from [RFC4328]. However a number of practical issues arise 1693 in the identification of wavelengths and signals, and distributed 1694 wavelength assignment processes which are discussed below. 1696 6.1.1. Identifying Wavelengths and Signals 1698 As previously stated a global fixed mapping between wavelengths and 1699 labels simplifies the characterization of WDM links and WSON devices. 1701 Furthermore such a mapping as described in [Otani] provides such a 1702 fixed mapping for communication between PCE and WSON PCCs. 1704 6.1.2. WSON Signals and Network Element Processing 1706 As discussed in Section 3.3.2. a WSON signal at any point along its 1707 path can be characterized by the (a) modulation format, (b) FEC, (c) 1708 wavelength, (d)bit rate, and (d)G-PID. 1710 Currently G-PID, wavelength (via labels), and bit rate (via bandwidth 1711 encoding) are supported in [RFC3471] and [RFC3473]. These RFCs can 1712 accommodate the wavelength changing at any node along the LSP and can 1713 thus provide explicit control of wavelength converters. 1715 In the fixed regeneration point scenario described in Section 5.5.1. 1716 (Fixed Regeneration Points) no enhancements are required to signaling 1717 since there are no additional configuration options for the LSP at a 1718 node. 1720 In the case of shared regeneration pools described in Section 5.5.2. 1721 (Shared Regeneration Pools) it is necessary to indicate to a node 1722 that it should perform regeneration on a particular signal. Viewed 1723 another way, for an LSP, it is desirable to specify that certain 1724 nodes along the path perform regeneration. Such a capability 1725 currently does not exist in GMPLS signaling. 1727 The case of configurable regenerators described in Section 5.5.3. 1728 (Reconfigurable Regenerators) is very similar to the previous except 1729 that now there are potentially many more items that can be configured 1730 on a per node basis for an LSP. 1732 Note that the techniques of [RFC5420] which allow for additional LSP 1733 attributes and their recording in a Record Route Object (RRO) object 1734 could be extended to allow for additional LSP attributes in an ERO. 1735 This could allow one to indicate where optional 3R regeneration 1736 should take place along a path, any modification of LSP attributes 1737 such as modulation format, or any enhance processing such as 1738 performance monitoring. 1740 6.1.3. Combined RWA/Separate Routing WA support 1742 In either the combined RWA or separate routing WA cases, the node 1743 initiating the signaling will have a route from the source to 1744 destination along with the wavelengths (generalized labels) to be 1745 used along portions of the path. Current GMPLS signaling supports an 1746 Explicit Route Object (ERO) and within an ERO an ERO Label subobject 1747 can be used to indicate the wavelength to be used at a particular 1748 node. In case the local label map approach is used the label sub- 1749 object entry in the ERO has to be interpreted appropriately. 1751 6.1.4. Distributed Wavelength Assignment: Unidirectional, No 1752 Converters 1754 GMPLS signaling for a unidirectional optical path LSP allows for the 1755 use of a label set object in the Resource Reservation Protocol - 1756 Traffic Engineering (RSVP-TE) path message. The processing of the 1757 label set object to take the intersection of available lambdas along 1758 a path can be performed resulting in the set of available lambda 1759 being known to the destination that can then use a wavelength 1760 selection algorithm to choose a lambda. 1762 6.1.5. Distributed Wavelength Assignment: Unidirectional, Limited 1763 Converters 1765 In the case of wavelength converters, nodes with wavelength 1766 converters would need to make the decision as to whether to perform 1767 conversion. One indicator for this would be that the set of available 1768 wavelengths which is obtained via the intersection of the incoming 1769 label set and the output links available wavelengths is either null 1770 or deemed too small to permit successful completion. 1772 At this point the node would need to remember that it will apply 1773 wavelength conversion and will be responsible for assigning the 1774 wavelength on the previous lambda-contiguous segment when the RSVP-TE 1775 RESV message is processed. The node will pass on an enlarged label 1776 set reflecting only the limitations of the wavelength converter and 1777 the output link. The record route option in RSVP-TE signaling can be 1778 used to show where wavelength conversion has taken place. 1780 6.1.6. Distributed Wavelength Assignment: Bidirectional, No 1781 Converters 1783 There are cases of a bidirectional optical path which requires the 1784 use of the same lambda in both directions. The above procedure can be 1785 used to determine the available bidirectional lambda set if it is 1786 interpreted that the available label set is available in both 1787 directions. In bidirectional LSPs setup, according to [RFC3471] 1788 Section 4.1. (Architectural Approaches to RWA), is indicated by the 1789 presence of an upstream label in the path message. 1791 However, until the intersection of the available label sets is 1792 determined along the path and at the destination node the upstream 1793 label information may not be correct. This case can be supported 1794 using current GMPLS mechanisms, but may not be as efficient as an 1795 optimized bidirectional single-label allocation mechanism. 1797 6.2. Implications for GMPLS Routing 1799 GMPLS routing [RFC4202] currently defines an interface capability 1800 descriptor for "lambda switch capable" (LSC) which can be used to 1801 describe the interfaces on a ROADM or other type of wavelength 1802 selective switch. In addition to the topology information typically 1803 conveyed via an IGP, it would be necessary to convey the following 1804 subsystem properties to minimally characterize a WSON: 1806 1. WDM Link properties (allowed wavelengths). 1808 2. Optical transmitters (wavelength range). 1810 3. ROADM/FOADM Properties (connectivity matrix, port wavelength 1811 restrictions). 1813 4. Wavelength converter properties (per network element, may change if 1814 a common limited shared pool is used). 1816 This information is modeled in detail in [WSON-Info] and a compact 1817 encoding is given in [WSON-Encode]. 1819 6.2.1. Electro-Optical Element Signal Compatibility 1821 In network scenarios where signal compatibility is a concern it is 1822 necessary to add parameters to our existing node and link models to 1823 take into account electro-optical input constraints, output 1824 constraints, and the signal processing capabilities of a NE in path 1825 computations. 1827 Input constraints: 1829 1. Permitted optical tributary signal classes: A list of optical 1830 tributary signal classes that can be processed by this network 1831 element or carried over this link. (configuration type) 1832 2. Acceptable FEC codes. (configuration type) 1833 3. Acceptable Bit Rate Set: a list of specific bit rates or bit rate 1834 ranges that the device can accommodate. Coarse bit rate info is 1835 included with the optical tributary signal class restrictions. 1836 4. Acceptable G-PID list: a list of G-PIDs corresponding to the 1837 "client" digital streams that is compatible with this device. 1839 Note that the bit rate of the signal does not change over the LSP. 1840 This can be communicated as an LSP parameter and hence this 1841 information would be available for any NE that needs to use it for 1842 configuration. Hence it is not necessary to have "configuration type" 1843 for the NE with respect to bit rate. 1845 Output constraints: 1847 1. Output modulation: (a)same as input, (b) list of available types 1849 2. FEC options: (a) same as input, (b) list of available codes 1851 Processing capabilities: 1853 1. Regeneration: (a) 1R, (b) 2R, (c) 3R, (d)list of selectable 1854 regeneration types 1856 2. Fault and performance monitoring: (a) G-PID particular 1857 capabilities, (b) optical performance monitoring capabilities. 1859 Note that such parameters could be specified on an (a) Network 1860 element wide basis, (b) a per port basis, (c) on a per regenerator 1861 basis. Typically such information has been on a per port basis; see 1862 the GMPLS interface switching capability descriptor [RFC4202]. 1864 6.2.2. Wavelength-Specific Availability Information 1866 For wavelength assignment it is necessary to know which specific 1867 wavelengths are available and which are occupied if a combined RWA 1868 process or separate WA process is run as discussed in sections 4.1.1. 1869 4.1.2. This is currently not possible with GMPLS routing. 1871 In the routing extensions for GMPLS [RFC4202], requirements for 1872 layer-specific TE attributes are discussed. RWA for optical networks 1873 without wavelength converters imposes an additional requirement for 1874 the lambda (or optical channel) layer: that of knowing which specific 1875 wavelengths are in use. Note that current DWDM systems range from 16 1876 channels to 128 channels with advanced laboratory systems with as 1877 many as 300 channels. Given these channel limitations and if the 1878 approach of a global wavelength to label mapping or furnishing the 1879 local mappings to the PCEs is taken then representing the use of 1880 wavelengths via a simple bit-map is feasible [Gen-Encode]. 1882 6.2.3. WSON Routing Information Summary 1884 The following table summarizes the WSON information that could be 1885 conveyed via GMPLS routing and attempts to classify that information 1886 as to its static or dynamic nature and whether that information would 1887 tend to be associated with either a link or a node. 1889 Information Static/Dynamic Node/Link 1890 ------------------------------------------------------------------ 1891 Connectivity matrix Static Node 1892 Per port wavelength restrictions Static Node(1) 1893 WDM link (fiber) lambda ranges Static Link 1894 WDM link channel spacing Static Link 1895 Optical transmitter range Static Link(2) 1896 Wavelength conversion capabilities Static(3) Node 1897 Maximum bandwidth per wavelength Static Link 1898 Wavelength availability Dynamic(4) Link 1899 Signal compatibility and processing Static/Dynamic Node 1901 Notes: 1903 1. These are the per port wavelength restrictions of an optical 1904 device such as a ROADM and are independent of any optical 1905 constraints imposed by a fiber link. 1907 2. This could also be viewed as a node capability. 1909 3. This could be dynamic in the case of a limited pool of converters 1910 where the number available can change with connection 1911 establishment. Note it may be desirable to include regeneration 1912 capabilities here since OEO converters are also regenerators. 1914 4. Not necessarily needed in the case of distributed wavelength 1915 assignment via signaling. 1917 While the full complement of the information from the previous table 1918 is needed in the Combined RWA and the separate Routing and WA 1919 architectures, in the case of Routing + distributed WA via signaling 1920 only the following information is needed: 1922 Information Static/Dynamic Node/Link 1923 ------------------------------------------------------------------ 1924 Connectivity matrix Static Node 1925 Wavelength conversion capabilities Static(3) Node 1927 Information models and compact encodings for this information is 1928 provided in [WSON-Info], [Gen-Encode] and [WSON-Encode]. 1930 6.3. Optical Path Computation and Implications for PCE 1932 As previously noted RWA can be computationally intensive. Such 1933 computationally intensive path computations and optimizations were 1934 part of the impetus for the PCE architecture [RFC4655]. 1936 The Path Computation Element Protocol (PCEP) defines the procedures 1937 necessary to support both sequential [RFC5440] and global concurrent 1938 path computations (PCE-GCO) [RFC5557]. The PCEP is well positioned to 1939 support WSON-enabled RWA computation with some protocol enhancement. 1941 Implications for PCE generally fall into two main categories: (a) 1942 optical path constraints and characteristics, (b) computation 1943 architectures. 1945 6.3.1. Optical path Constraints and Characteristics 1947 For the varying degrees of optimization that may be encountered in a 1948 network the following models of bulk and sequential optical path 1949 requests are encountered: 1951 o Batch optimization, multiple optical paths requested at one time 1952 (PCE-GCO). 1954 o Optical path(s) and backup optical path(s) requested at one time 1955 (PCEP). 1957 o Single optical path requested at a time (PCEP). 1959 PCEP and PCE-GCO can be readily enhanced to support all of the 1960 potential models of RWA computation. 1962 Optical path constraints include: 1964 o Bidirectional Assignment of wavelengths. 1966 o Possible simultaneous assignment of wavelength to primary and 1967 backup paths. 1969 o Tuning range constraint on optical transmitter. 1971 6.3.2. Electro-Optical Element Signal Compatibility 1973 When requesting a path computation to PCE, the PCC should be able to 1974 indicate the following: 1976 o The G-PID type of an LSP. 1978 o The signal attributes at the transmitter (at the source): (i) 1979 modulation type; (ii) FEC type. 1981 o The signal attributes at the receiver (at the sink): (i) 1982 modulation type; (ii) FEC type. 1984 The PCE should be able to respond to the PCC with the following: 1986 o The conformity of the requested optical characteristics associated 1987 with the resulting LSP with the source, sink and NE along the LSP. 1989 o Additional LSP attributes modified along the path (e.g., 1990 modulation format change, etc.). 1992 6.3.3. Discovery of RWA Capable PCEs 1994 The algorithms and network information needed for RWA are somewhat 1995 specialized and computationally intensive hence not all PCEs within a 1996 domain would necessarily need or want this capability. Hence, it 1997 would be useful via the mechanisms being established for PCE 1998 discovery [RFC5088] to indicate that a PCE has the ability to deal 1999 with RWA. Reference [RFC5088] indicates that a sub-TLV could be 2000 allocated for this purpose. 2002 Recent progress on objective functions in PCE [RFC5541] would allow 2003 the operators to flexibly request differing objective functions per 2004 their need and applications. For instance, this would allow the 2005 operator to choose an objective function that minimizes the total 2006 network cost associated with setting up a set of paths concurrently. 2007 This would also allow operators to choose an objective function that 2008 results in a most evenly distributed link utilization. 2010 This implies that PCEP would easily accommodate wavelength selection 2011 algorithm in its objective function to be able to optimize the path 2012 computation from the perspective of wavelength assignment if chosen 2013 by the operators. 2015 7. Security Considerations 2017 This document has no requirement for a change to the security models 2018 within GMPLS and associated protocols. That is the OSPF-TE, RSVP-TE, 2019 and PCEP security models could be operated unchanged. 2021 However satisfying the requirements for RWA using the existing 2022 protocols may significantly affect the loading of those protocols. 2023 This may make the operation of the network more vulnerable to denial 2024 of service attacks. Therefore additional care maybe required to 2025 ensure that the protocols are secure in the WSON environment. 2027 Furthermore the additional information distributed in order to 2028 address RWA represents a disclosure of network capabilities that an 2029 operator may wish to keep private. Consideration should be given to 2030 securing this information. For a general discussion on MPLS and GMPLS 2031 related security issues, see the MPLS/GMPLS security framework 2032 [RFC5920]. 2034 8. IANA Considerations 2036 This document makes no request for IANA actions. 2038 9. Acknowledgments 2040 The authors would like to thank Adrian Farrel for many helpful 2041 comments that greatly improved the contents of this draft. 2043 This document was prepared using 2-Word-v2.0.template.dot. 2045 10. References 2047 10.1. Normative References 2049 [RFC3471] Berger, L., "Generalized Multi-Protocol Label Switching 2050 (GMPLS) Signaling Functional Description", RFC 3471, 2051 January 2003. 2053 [RFC3473] Berger, L., Ed., "Generalized Multi-Protocol Label 2054 Switching (GMPLS) Signaling Resource ReserVation Protocol- 2055 Traffic Engineering (RSVP-TE) Extensions", RFC 3473, 2056 January 2003. 2058 [RFC3945] Mannie, E. "Generalized Multi-Protocol Label Switching 2059 (GMPLS) Architecture", RFC 3945, October 2004. 2061 [RFC4202] Kompella, K. and Y. Rekhter, "Routing Extensions in Support 2062 of Generalized Multi-Protocol Label Switching (GMPLS)", RFC 2063 4202, October 2005. 2065 [RFC4328] Papadimitriou, D., "Generalized Multi-Protocol Label 2066 Switching (GMPLS) Signaling Extensions for G.709 Optical 2067 Transport Networks Control", RFC 4328, January 2006. 2069 [RFC4655] Farrel, A., Vasseur, JP., and Ash, J., "A Path Computation 2070 Element (PCE)-Based Architecture ", RFC 4655, August 2006. 2072 [RFC5088] J.L. Le Roux, J.P. Vasseur, Yuichi Ikejiri, and Raymond 2073 Zhang, "OSPF protocol extensions for Path Computation 2074 Element (PCE) Discovery", January 2008. 2076 [RFC5212] Shiomoto, K., Papadimitriou, D., Le Roux, JL., Vigoureux, 2077 M., and D. Brungard, "Requirements for GMPLS-Based Multi- 2078 Region and Multi-Layer Networks (MRN/MLN)", RFC 5212, July 2079 2008. 2081 [RFC5557] Y. Lee, J.L. Le Roux, D. King, and E. Oki, "Path 2082 Computation Element Communication Protocol (PCECP) 2083 Requirements and Protocol Extensions In Support of Global 2084 Concurrent Optimization", RFC 5557, July 2009. 2086 [RFC5420] Farrel, A., Ed., Papadimitriou, D., Vasseur, JP., and A. 2087 Ayyangarps, "Encoding of Attributes for MPLS LSP 2088 Establishment Using Resource Reservation Protocol Traffic 2089 Engineering (RSVP-TE)", RFC 5420, February 2009. 2091 [RFC5440] J.P. Vasseur and J.L. Le Roux (Editors), "Path Computation 2092 Element (PCE) Communication Protocol (PCEP)", RFC 5440, May 2093 2009. 2095 [RFC5541] J.L. Le Roux, J.P. Vasseur, and Y. Lee, "Encoding of 2096 Objective Functions in Path Computation Element (PCE) 2097 communication and discovery protocols", RFC 5541, July 2098 2009. 2100 10.2. Informative References 2102 [Gen-Encode] G. Bernstein, Y. Lee, D. Li, and W. Imajuku, "General 2103 Network Element Constraint Encoding for GMPLS Controlled 2104 Networks", draft-ietf-ccamp-general-constraint-encode, work 2105 in progress. 2107 [G.652] ITU-T Recommendation G.652, Characteristics of a single-mode 2108 optical fibre and cable, June 2005. 2110 [G.653] ITU-T Recommendation G.653, Characteristics of a dispersion- 2111 shifted single-mode optical fibre and cable, December 2006. 2113 [G.654] ITU-T Recommendation G.654, Characteristics of a cut-off 2114 shifted single-mode optical fibre and cable, December 2006. 2116 [G.655] ITU-T Recommendation G.655, Characteristics of a non-zero 2117 dispersion-shifted single-mode optical fibre and cable, 2118 March 2006. 2120 [G.656] ITU-T Recommendation G.656, Characteristics of a fibre and 2121 cable with non-zero dispersion for wideband optical 2122 transport, December 2006. 2124 [G.671] ITU-T Recommendation G.671, Transmission characteristics of 2125 optical components and subsystems, January 2005. 2127 [G.694.1] ITU-T Recommendation G.694.1, "Spectral grids for WDM 2128 applications: DWDM frequency grid", June, 2002. 2130 [G.872] ITU-T Recommendation G.872, Architecture of optical 2131 transport networks, November 2001. 2133 [G.959.1] ITU-T Recommendation G.959.1, Optical Transport Network 2134 Physical Layer Interfaces, March 2006. 2136 [G.694.1] ITU-T Recommendation G.694.1, Spectral grids for WDM 2137 applications: DWDM frequency grid, June 2002. 2139 [G.694.2] ITU-T Recommendation G.694.2, Spectral grids for WDM 2140 applications: CWDM wavelength grid, December 2003. 2142 [G.Sup39] ITU-T Series G Supplement 39, Optical system design and 2143 engineering considerations, February 2006. 2145 [G.Sup43] ITU-T Series G Supplement 43, Transport of IEEE 10G base-R 2146 in optical transport networks (OTN), November 2006. 2148 [Imajuku] W. Imajuku, Y. Sone, I. Nishioka, S. Seno, "Routing 2149 Extensions to Support Network Elements with Switching 2150 Constraint", work in progress: draft-imajuku-ccamp-rtg- 2151 switching-constraint. 2153 [Otani] T. Otani, H. Guo, K. Miyazaki, D. Caviglia, "Generalized 2154 Labels of Lambda-Switching Capable Label Switching Routers 2155 (LSR)", work in progress: draft-ietf-ccamp-gmpls-g-694- 2156 lambda-labels, work in progress. 2158 [RFC5920] Fang, L., "Security Framework for MPLS and GMPLS 2159 Networks", RFC 5920, July 2010.[Otani]T. Otani, H. Guo, K. 2160 Miyazaki, D. Caviglia, "Generalized Labels of Lambda- 2161 Switching Capable Label Switching Routers (LSR)", work in 2162 progress: draft-otani-ccamp-gmpls-g-694-lambda-labels, work 2163 in progress. 2165 [WSON-Encode] G. Bernstein, Y. Lee, D. Li, and W. Imajuku, "Routing 2166 and Wavelength Assignment Information Encoding for 2167 Wavelength Switched Optical Networks", draft-ietf-ccamp- 2168 rwa-wson-encode, work in progress. 2170 [WSON-Imp] Y. Lee, G. Bernstein, D. Li, G. Martinelli, "A Framework 2171 for the Control of Wavelength Switched Optical Networks 2172 (WSON) with Impairments", draft-ietf-ccamp-wson- 2173 impairments, work in progress. 2175 [WSON-Info] Y. Lee, G. Bernstein, D. Li, W. Imajuku, "Routing and 2176 Wavelength Assignment Information for Wavelength Switched 2177 Optical Networks", draft-bernstein-ccamp-wson-info, work in 2178 progress 2180 11. Contributors 2182 Snigdho Bardalai 2183 Fujitsu 2185 Email: Snigdho.Bardalai@us.fujitsu.com 2187 Diego Caviglia 2188 Ericsson 2189 Via A. Negrone 1/A 16153 2190 Genoa Italy 2192 Phone: +39 010 600 3736 2193 Email: diego.caviglia@(marconi.com, ericsson.com) 2195 Daniel King 2196 Old Dog Consulting 2197 UK 2199 Email: daniel@olddog.co.uk 2201 Itaru Nishioka 2202 NEC Corp. 2203 1753 Simonumabe, Nakahara-ku 2204 Kawasaki, Kanagawa 211-8666 2205 Japan 2207 Phone: +81 44 396 3287 2208 Email: i-nishioka@cb.jp.nec.com 2210 Lyndon Ong 2211 Ciena 2213 Email: Lyong@Ciena.com 2215 Pierre Peloso 2216 Alcatel-Lucent 2217 Route de Villejust, 91620 Nozay 2218 France 2220 Email: pierre.peloso@alcatel-lucent.fr 2222 Jonathan Sadler 2223 Tellabs 2224 Email: Jonathan.Sadler@tellabs.com 2226 Dirk Schroetter 2227 Cisco 2228 Email: dschroet@cisco.com 2230 Jonas Martensson 2231 Acreo 2232 Electrum 236 2233 16440 Kista, Sweden 2235 Email:Jonas.Martensson@acreo.se 2237 Author's Addresses 2239 Greg M. Bernstein (ed.) 2240 Grotto Networking 2241 Fremont California, USA 2243 Phone: (510) 573-2237 2244 Email: gregb@grotto-networking.com 2246 Young Lee (ed.) 2247 Huawei Technologies 2248 1700 Alma Drive, Suite 100 2249 Plano, TX 75075 2250 USA 2252 Phone: (972) 509-5599 (x2240) 2253 Email: ylee@huawei.com 2255 Wataru Imajuku 2256 NTT Network Innovation Labs 2257 1-1 Hikari-no-oka, Yokosuka, Kanagawa 2258 Japan 2260 Phone: +81-(46) 859-4315 2261 Email: imajuku.wataru@lab.ntt.co.jp 2263 Intellectual Property Statement 2265 The IETF Trust takes no position regarding the validity or scope of 2266 any Intellectual Property Rights or other rights that might be 2267 claimed to pertain to the implementation or use of the technology 2268 described in any IETF Document or the extent to which any license 2269 under such rights might or might not be available; nor does it 2270 represent that it has made any independent effort to identify any 2271 such rights. 2273 Copies of Intellectual Property disclosures made to the IETF 2274 Secretariat and any assurances of licenses to be made available, or 2275 the result of an attempt made to obtain a general license or 2276 permission for the use of such proprietary rights by implementers or 2277 users of this specification can be obtained from the IETF on-line IPR 2278 repository at http://www.ietf.org/ipr 2280 The IETF invites any interested party to bring to its attention any 2281 copyrights, patents or patent applications, or other proprietary 2282 rights that may cover technology that may be required to implement 2283 any standard or specification contained in an IETF Document. Please 2284 address the information to the IETF at ietf-ipr@ietf.org. 2286 Disclaimer of Validity 2288 All IETF Documents and the information contained therein are provided 2289 on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE 2290 REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE 2291 IETF TRUST AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL 2292 WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY 2293 WARRANTY THAT THE USE OF THE INFORMATION THEREIN WILL NOT INFRINGE 2294 ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS 2295 FOR A PARTICULAR PURPOSE. 2297 Acknowledgment 2299 Funding for the RFC Editor function is currently provided by the 2300 Internet Society.