idnits 2.17.1 draft-kalevi-simple-media-access-01.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** Cannot find the required boilerplate sections (Copyright, IPR, etc.) in this document. Expected boilerplate is as follows today (2024-04-24) according to https://trustee.ietf.org/license-info : IETF Trust Legal Provisions of 28-dec-2009, Section 6.a: This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. IETF Trust Legal Provisions of 28-dec-2009, Section 6.b(i), paragraph 2: Copyright (c) 2024 IETF Trust and the persons identified as the document authors. All rights reserved. IETF Trust Legal Provisions of 28-dec-2009, Section 6.b(i), paragraph 3: This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** Missing expiration date. The document expiration date should appear on the first and last page. ** The document seems to lack a 1id_guidelines paragraph about Internet-Drafts being working documents. ** The document seems to lack a 1id_guidelines paragraph about the list of current Internet-Drafts. ** The document seems to lack a 1id_guidelines paragraph about the list of Shadow Directories. == No 'Intended status' indicated for this document; assuming Proposed Standard == It seems as if not all pages are separated by form feeds - found 0 form feeds but 24 pages Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) ** There are 4 instances of too long lines in the document, the longest one being 1 character in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == Line 94 has weird spacing: '...service speci...' == Line 355 has weird spacing: '...User charging...' == Line 434 has weird spacing: '...User input to...' == Line 455 has weird spacing: '...scarded tra...' == Line 572 has weird spacing: '... load effor...' == (6 more instances...) -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (19 December 1997) is 9623 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) No issues found here. Summary: 7 errors (**), 0 flaws (~~), 8 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 Network Working group K. Kilkki 2 Nokia Research Center 3 Internet-Draft June 1997 4 Expire in 19 December 1997 6 Simple Integrated Media Access (SIMA) 8 Status of this Memo 10 This document is an Internet-Draft. Internet-Drafts are working 11 documents of the Internet Engineering Task Force (IETF), its areas, and 12 its working groups. Note that other groups may also distribute working 13 documents as Internet-Drafts. 15 Internet-Drafts are draft documents valid for a maximum of six months 16 and may be updated, replaced, or obsoleted by other documents at any 17 time. It is inappropriate to use Internet-Drafts as reference material 18 or to cite them other than as "work in progress." 20 To learn the current status of any Internet-Draft, please check the 21 "1id-abstracts.txt" listing contained in the Internet-Drafts Shadow 22 Directories on ftp.is.co.za (Africa), ftp.nordu.net (Europe), 23 munnari.oz.au (Pacific Rim), ds.internic.net (US East Coast), 24 orftp.isi.edu (US West Coast). 26 Distribution of this memo is unlimited. 27 Please send comments to kalevi.kilkki@research.nokia.com 29 Abstract 31 The basic objectives of future Internet are to increase the network 32 capacity, to offer a practical real-time service, and to develop a 33 feasible charging scheme. These objectives introduce very strict 34 requirements for the traffic control system. This document presents a 35 new simple approach for traffic management: Simple Integrated Media 36 Access (SIMA) service. According to the SIMA concept each customer shall 37 define only two issues before a connection establishment: a nominal bit 38 rate (NBR) and the selection between real-time and non-real-time service 39 classes. NBR has two purposes: it forms the basis of charging, and it 40 defines how the network capacity is divided among different connections 41 during overload situations. Simplicity means that, on the one hand, the 42 network operator does not guarantee the continuous availability of 43 nominal bit rate, and on the other hand, the user is allowed to send 44 data with any bit rate independently of the NBR. However, the bit rate 45 of transmission defines the cell loss probability in the case of network 46 congestion. In this way a simple, but effective, self-regulation of 47 traffic can be realized. The strength of SIMA lies in its wide area of 48 applications. There is no need to build complex systems with several 49 service classes each appropriate to only certain applications. 51 Table of Contents 53 Abstract 1 54 1. Introduction 2 55 2. End-to-End Behavior 3 56 2.1 Nominal Bit Rate 4 57 2.2 Real-time and non-real-time services 4 58 2.3 Quality of the SIMA service 5 59 2.4 SIMA service chain 7 60 3. Motivation 9 61 3.1 Current services 9 62 3.2 Service comparison 11 63 3.3 Management comparison 13 64 3.4 Performance comparison 15 65 4. Network Element Data Handling Requirements 15 66 4.1 Access Node 16 67 4.2 Scheduling and buffering in core network 17 68 5. Invocation Information 18 69 6. Exported Information 19 70 7. Policing 19 71 8. Ordering and Merging 19 72 9. Guidelines for Implementors 19 73 9.1 Actual bit rate measurement 19 74 9.2 Implementation of scheduling algorithm 21 75 9.3 ATM implementation 21 76 9.4 The location of priority bits 22 77 10. Evaluation Criteria 22 78 11. Examples of Implementation 23 79 12. Examples of Use 23 80 13. Security Considerations 24 81 Author's address 24 82 Expiration 24 84 1. Introduction 86 The Internet is at a phase of great changes. There are several stringent 87 new requirements for the network because of two reasons: the invasion of 88 new users, and the rapid development of new applications. These 89 requirements mean that network capacity must rapidly be increased, real- 90 time service has to be fundamentally improved, and a feasible charging 91 scheme must be introduced. 93 The current Internet approach for meeting these requirements consists of 94 several service specifications, Resource Reservation Protocol, QoS 95 routing, etc. We can make an interpretation on the basis of the service 96 specifications that the basic philosophy of Internet development is to 97 define different services for different basic communication needs. There 98 seems to be demand for three elementary services: first one for very 99 reliable and high quality connections, second for connections with less 100 stringent quality requirements, and third one for data connections which 101 can smoothly adapt their bit rate. As the requirements of these 102 elementary classes differ significantly, an obvious approach is to have 103 different service specifications like the guaranteed service, 104 controlled-load service and best effort service specified in IETF drafts 105 (note that this approach has been used in the case of ATM as well). 107 The supposed advantage of this approach is that by dividing the service 108 specification task into several smaller parts the specifications process 109 is easier than if the all the service types were included in the one 110 specification. However, this advantage is somewhat questionable because 111 the whole service concept (with all the different service types) is what 112 the network operator should manage and sell to customers and what the 113 customer should buy and use. In particular, most Internet customers will 114 be reluctant to learn several complicated services which may even have 115 very different structures, traffic parameters, charging schemes, etc. 116 For real marketing purposes, the Internet service package must be 117 simple, much simpler than what the current service models directly make 118 possible. There are two main approaches to meet this requirement: to 119 hide the complexity of network service from the end-user or to design an 120 entirely new integrated service which is able to satisfy all the primary 121 customer needs. The approach used in this document belongs to the later 122 category. 124 2. End-to-End Behavior 126 The primary idea of the SIMA service is to maximize the exploitation of 127 network resources with a simple control scheme while keeping the ratios 128 of QoS levels offered to different flows unchanged under changeable 129 traffic conditions. The maximization is based on three key properties of 130 traffic control: all flows with different QoS requirements share the 131 total capacity of every link, the network attempts to avoid any 132 unnecessary packet discarding, and flow (or call) level blocking can be 133 avoided. The approximate constancy of QoS ratios and simplicity are 134 achieved by using 8 priority levels which make possible a fair packet 135 discarding scheme inside the network without keeping track on the 136 traffic of every flow. 138 The SIMA specification covers the whole Internet service including 139 charging, QoS and performance aspects, and traffic control functions in 140 the network. As opposed to most service specifications, charging is the 141 starting point of the SIMA concept. The prevalent charging scheme 142 applied by Internet operators is a flat rate one with a constant monthly 143 fee. Although this scheme is most reasonable when the network service is 144 based on the best-effort principle, many network operators may still be 145 willing to apply this scheme even with more complicated service models. 146 The SIMA service model is able to meet this demand. 148 2.1 Nominal Bit Rate 150 Shortly, with the SIMA service scheme, first a customer pays for some 151 Nominal Bit Rate (NBR, kbit/s) and then he/she can trade the speed for 152 QoS. Let us assume that a user pays $X/month. This charge is translated 153 to a Nominal Bit Rate using an arbitrary function. The function NBR = 154 F(X) could be linear, but there is no reason to specify the relationship 155 between NBR and charging. If NBR is permanent, it can be related to an 156 interface (we may assume the same organization that owns the right to 157 use a network interface buys the NBR). The next level of NBR is the NBR 158 assigned to a user (or IP-address). The bottom level is the NBR of a 159 flow (determined for instance by a pair of IP address and port number). 161 Depending on the available information and the network capabilities, 162 there are three basic approaches to manage NBRs. The simplest approach 163 is to assign the NBR only to an interface, which means that the network 164 measures the whole traffic going through the interface and handles this 165 traffic as an indivisible entity. The users and flows that share the NBR 166 obtain approximately the same QoS. In the second approach each user 167 (identified by an IP address) has his/her own NBR. Now the network 168 measures the total traffic generated by a user, and different flows 169 compete with each other on a best-effort basis. 171 Both these approaches have the serious drawback that they do not 172 separate different applications properly: a high-speed file transfer may 173 disturb other flows, e.g., real time video connections, although the 174 user may consider the file transfer as a background process which uses 175 only the capacity left by other more demanding applications. Therefore, 176 as regards the performance and QoS of the SIMA service the most useful 177 approach is the one where every flow has its own NBR. Later in this 178 draft we suppose that the network is capable to identify and measure 179 every flow, and that every flow has its own NBR. The question how these 180 NBRs are determined and managed can be left for network operators, and 181 is, therefore, out of the scope of this draft. 183 2.2 Real-time and non-real-time services 185 The other part of the SIMA service concept is the possibility to request 186 a real-time service. The user is entitled to him/herself determine 187 whether the flow is a real-time (rt) or non-real-time (nrt) one. In 188 practice, this decision can be made usually at the application level: a 189 real-time service is usually requested only for interactive audio or 190 video applications. If a real-time service is requested, the SIMA 191 network attempts to offer as short delay and small delay variation as 192 possible. The expense of this choice is that, if there are traffic 193 variations of time scale from 0.1 ms to 10 ms, small real-time buffers 194 cannot filter these variations. Therefore, the measurement for the 195 priority determination shall be more sensible as regards the traffic 196 variations in case of real-time service than with non-real-time service 197 (see illustration in Fig. 1). 199 If the user changes a VBR connection from nrt-service to rt-service 200 without changing NBR or traffic process, the delay will decrease, but 201 the cell loss ratio may increase because the real-time measurement gives 202 worse priorities during peak rates. If the user wants to obtain the same 203 quality, this impairment of loss ratio should be compensated by 204 increasing NBR. Real-time-service could, in this respect, be more 205 expensive than non-real-time service although there is no difference in 206 the tariffs. 208 In consequence, if the application is a real-time one, it is 209 advantageous for the user to select the real-time class, because it is 210 the only way to attain small delay and delay variation. Furthermore, if 211 the traffic variations are small enough, the user may always select a 212 real-time service, because there is no difference in cell loss ratio 213 between rt and nrt-services. In contrast, if there are significant 214 traffic variations as with typical data applications, the non-real-time 215 service gives better quality, that is, smaller packet loss ratio. 217 | * = actual bit rate | 218 | o = real-time measured bit rate | 219 |******o*o*o*o* x = nrt measured bit rate | 220 | oo | 221 | o | 222 | o | 223 | o o ****o*o*o*o*o*o*o*o*o | 224 | oo | 225 | o o | 226 |o xxxxxxoxxxxxxx o xxxxxxxxxxxxxxxxx | 227 | xxx oo xxxxxxxxxxxxxxxxxxxxx o xxxx| 228 | xxx ******o*o*o*o*o*o*o*o*o oo | 229 | xx ***o*o*o| 230 |xx | 231 o------------------------------------------------------------------- 233 Fig. 1. The difference between actual bit rate, measured bit rate for a 234 real-time flow and measured bit rate for a non-real-time flow. 236 2.3 Quality of the SIMA service 238 The total SIMA service requested by a user consists of a nominal bit 239 rate and of a possible real-time service request. This half of the 240 service is clear and reasonable. The other half of the service is the 241 expected QoS of the flow, or actually, the expected QoS of the 242 application that the customer uses over the SIMA network. An essential 243 issue for the success of the SIMA service is how reasonable and 244 acceptable this part of the service concept will be. 246 Most customers have experience of circuit switched networks (like 247 telephone networks) and packet networks with best-effort service (like 248 the current Internet). In a circuit switched network a busy period means 249 that the call blocking probability increases, which means that the 250 quality perceived by some customers drop occasionally to zero. In packet 251 networks the packet loss ratio increases during busy periods, and 252 effectively, the available capacity for a flow decreases if a TCP/IP 253 type of protocol is used. In a SIMA environment, when a user buys a NBR 254 for a flow and then sends traffic into a SIMA network, there is usually 255 no flow level blocking (although it is possible to protect the SIMA 256 network from excessive overloads by restricting the total sum of NBRs). 257 The quality of the flow depends on two issues: the NBR to actual bit rat 258 ratio, and total load in the network. 260 Therefore, a potential difficulty with the SIMA service is that the 261 customer cannot precisely know what the QoS of a flow will be because 262 rapid traffic variations may bring about unexpected changes of QoS. 263 However, even in the case of services using resource reservation the 264 actual quality of flows using certain quality class may vary 265 significantly, because the quality can only be determined by using 266 statistical parameters. Furthermore, as the resource reservation 267 principle may result in flow level blocking, a high quality connection 268 cannot be guaranteed during overload situations. 270 Because the quality of existing flows is not in the same way predictable 271 as with services using a complicated resource reservation mechanism, the 272 SIMA network shall be implemented in a way that the users can rely on 273 the fairness of the service. The fairness of the SIMA service is based 274 on the fact that all flows with the same actual bit rate to NBR ratio 275 perceives similar QoS. Thus, a home user with 10 Kbit/s NBR receives the 276 same QoS as a large company with NBR of 100 Mbit/s provided that both 277 are transmitting at their own NBR. The SIMA service can offer this 278 fairness feature during a short interval. In contrast, during long 279 period, like a month, fairness is not as clear if flat rate scheme is 280 applied, because the amount of transferred information depends 281 essentially on total length of active periods, whereas the charging does 282 not depend on the activity of customer. This fairness problem common to 283 any service with flat-rate charging can be solved, if needed, by using a 284 time dependent charging scheme. 286 Another aspect of fairness is the possibility to obtain more quality 287 with higher price or lower price with less quality by changing the 288 actual bit rate or NBR. This means that each customer is entitled to 289 change the NBR to actual bit rate ratio (b) and by that means to 290 optimize his/her quality to charge ratio. If the ratio increases, the 291 quality of the flow is enhanced. If the user sends traffic by using a 292 constant bit rate, the SIMA service offers 7 different quality levels 293 (for variable bit rate traffic the levels are more distinct but 294 basically the same). Although the absolute quality of each class depends 295 on the network dimensioning and on actual traffic process, the quality 296 levels can be described approximately as follows: 298 7 = reserved for non-SIMA services with resource reservation 299 6 = excellent quality: negligible packet loss ratio 300 5 = high quality: packet losses only during exceptional traffic peaks 301 4 = good quality: small packet loss ratio even during busy hour 302 3 = moderate quality: usually small packet loss ratio except during busy 303 hours 304 2 = satisfactory quality: from time to time very high packet loss ratio 305 1 = suitable for best-effort traffic during busy hour 306 0 = unusable during busy hour, but suitable for best-effort traffic 307 during non-busy hours 309 The charge of priority level j will be X*2^(j-4), if the charge of level 310 4 is X, and if the charging is proportional to NBR. However, quality 311 level 0 can be in practice obtained free of charge. The network operator 312 may try to dimension the network in a way that the traffic of 3 lowest 313 levels is able to fill the network capacity left by the higher priority 314 levels during busy hour. As the charge of level 6 service is 16 times 315 higher than that of level 2, we can assume that there will be much more 316 traffic offered to the lowest levels. For instance, there could be 10 317 times more traffic on the lowest levels and still the incomes from level 318 6 traffic is higher than those of levels 0, 1, and 2 together. 319 Therefore, the traffic load of level 6 could be increased by several 320 hundreds percents before there is any packet loss. Note that even very 321 high load of the low quality levels has no significant effect on the 322 packet loss ratio of the higher levels. It is reasonable to assume that 323 the most intense traffic variations occur at the lowest quality levels, 324 whereas the charging may dampen the variations at the highest quality 325 levels. Thus, for most of the time higher priority levels can be 326 considered as insulated from the lower levels having varying packet loss 327 ratio. 329 2.4 SIMA service chain 331 As a conclusion the SIMA service chain can be outlined as in Fig 2. The 332 user input to the SIMA network consists of charge (C -> X $), actual 333 traffic sent into the network (T), and rt/nrt selection (the only 334 traffic or quality parameter). The network may inform the user of the 335 offered service tariffs by announcing the NBR (or as the user is able to 336 know the charging function, he may select directly a proper NBR). The 337 main output of the network is the actual QoS of the flow, which depends 338 directly on the network performance. Note that there is a (one-way) 339 connection from the charge of the flow to the actual QoS of the flow via 340 NBR, the SIMA control and network performance. Therefore, although there 341 is no pre-defined exact relation between charging and QoS, the user may 342 optimize the charging of the flow by trying firstly a low charge, and 343 then doubling the charge until the quality level is sufficient (or when 344 only flat rate charging is used, the user may change the actual bit rate 345 of the flow). 347 Another important feature is that the traffic control information is 348 conveyed purely by the SIMA packets (or cells), which means that there 349 is no need to have any separate control information transported between 350 different network nodes. Finally, as there is no packet or cell 351 discarding based on a separate traffic flow, packets or cells are 352 discarded only if the total load exceeds the networks capacity. In this 353 way the total capacity of the network can be exploited very efficiently 355 User charging input to traffic performance 356 network control 358 C --> X $ 359 + 360 tariff 361 function => NBR for the flow 362 I <- - - - - - -/ + 363 T --------------> actual offered 364 traffic of 365 the flow 366 + 367 P --------------> rt/nrt => SIMA packets 368 + 369 SIMA traffic control 370 + 371 aggregate traffic 372 process 373 + 374 network capacity => network 375 Q <---------------------------------------------------- performance 377 Fig. 2. Service chain of SIMA. C is the user's readiness to pay, I is 378 information given by the network, T is actual traffic sent into the 379 network, 380 P is the parameters needed to control the flow, and 381 Q is the quality experienced by the user. 383 3. Motivation 385 3.1 Current services 387 The SIMA service is intended to give a reasonable service concept for 388 ordinary Internet users while offering quality and fairness. The main 389 motivation of the SIMA service concept is the apparent unsuitability of 390 current Internet service concepts to entirely meet these requirements. 391 The starting point of the development of Internet services is the best- 392 effort service. The best-effort service chain is presented in Fig. 3. 394 The well-known problems of best-effort service are that there is no 395 relation between quality and charging, and that there is no way to offer 396 high quality (small packet loss or small delay) for those flows that 397 need these features. The prevalent approach to solve this problem is to 398 design guaranteed service classes, each of which has certain quality 399 features. A simplified service chain of this approach is presented in 400 Fig. 4. 402 User traffic performance 403 control 405 C <--- charging X $ 406 ===================================== 408 T -------------------> offered 409 traffic 410 + 411 aggregate 412 traffic process 413 + 414 FIFO buffers 415 + 416 network 417 capacity => network 418 Q <------------------------------------ performance 420 Fig. 3. Service chain of a best-effort service 422 The user input to a network with a guaranteed service consists of 423 requested traffic and quality parameters (P), and actual traffic sent 424 into the network (T). Actually, this is a significant difference between 425 SIMA and guaranteed service. With SIMA a customer mainly informs how 426 much he/she is willing to pay. With guaranteed service the customer must 427 first predict the parameters of his/her flow, something that is not easy 428 even for an expert. The network informs users of the charge of the flow 429 by using a complicated tariff table including all possible combinations 430 of traffic and quality parameters. Because there will always be a lot of 431 customers that are not willing to make familiar with these parameters, 432 there must be some means to hide the complexity of the service. 434 User input to traffic performance 435 network control 437 C <-------------------------------- 438 | 439 I <- - - - - - - - - - - | 440 | | 441 P --> requested | + tariff => charging X $ 442 traffic | table 443 + | 444 quality | 445 parameters | => service class 446 + 447 traffic 448 descript. => "flow + other control 449 control" functions 450 + + 451 T --------------------------------> offered => accepted 452 traffic traffic 453 | + 454 V aggregate 455 discarded traffic 456 Q1 <------------------------------- traffic process 457 + 458 network 459 capacity => network 460 Q2 <----------------------------------------------------------- 461 performance 463 Fig. 4. Service chain of a guaranteed service 465 The guaranteed service approach means that the network attempts to give 466 a statistical prediction of the actual quality of the flow: certain 467 service class will generate certain average quality (it is assumed that 468 each user is willing and able to understand the meaning of the quality 469 parameters). However, because of the variations in aggregate traffic 470 process the actual quality can sometimes be worse than predicted, but in 471 great majority of cases if will be much better. Therefore, the 472 connection between requested quality and actual quality is more 473 complicated than can be concluded directly from the service 474 specifications. 476 The output of the network is the actual QoS of the flow, but in this 477 case the quality (or rather, quality impairments) consists of two parts. 478 First one containing the possible packet loss ratio due to the control 479 of each flow ("flow control" in Fig. 4, UPC in ATM networks), and the 480 other one containing the effects of control functions directed to the 481 aggregate traffic load. In order to be able to respond properly in a 482 case of insufficient quality, these effects have to be discerned since 483 either the traffic parameters or service class should be changed. 484 Therefore, although the user can be able to optimize the quality to 485 charge ratio by changing traffic and/or quality parameters, this 486 optimization needs quite profound understanding of the properties of 487 services, network and traffic (or a very intelligent application to 488 perform the task). 490 3.2 Service comparison 492 A serious difficulty of most services with guaranteed quality is how to 493 build a reasonable service package offered to ordinary customers not 494 familiar with technical details. When comparing the SIMA service with 495 other possible integrated service concepts, the principal question is 496 whether the SIMA service can in this respect be better than the other 497 approaches. It is important to note that it is not reasonable to only 498 compare individual services realized by SIMA, or some other service 499 models, but the whole service package offered to customers. 501 Table 1 provides a brief summary for the comparison. The prevalent 502 approaches are service specifications developed at IETF's Integrated 503 Service working group (IntServ), and ATM specifications. We may assume 504 that most future Internet customers have different service requirements. 505 The two main services needed are a real-time service with high quality, 506 and a file transfer service with loose requirements for packet loss 507 ratio and delay. In addition, some customers may benefit from a service 508 which guarantees a small packet loss ratio but does not provide small 509 delay. If the network operator attempts to satisfy all these 510 requirements by using the current specifications, he must implement 511 several services. Possible combinations are: guaranteed service, 512 controlled load service and best-effort service if IETF's specifications 513 are used, and CBR+rt-VBR, ABR and UBR if ATM is used. 515 For ordinary customers charging has to be understandable, acceptable, 516 reliable, believable, etc. These properties cannot be attained if the 517 charging structure is too complicated. One of the main problems with the 518 current approaches is that the whole service offered to customers 519 consists of several services which may, and likely will, have different 520 charging principles. The charging of any service based on resource 521 reservations is likely to be based on the traffic and QoS parameters 522 used at the reservation phase. Best-effort service cannot apply the same 523 scheme as there is no reservation, instead, flat rate or usage based 524 charging schemes are usable. The charging of controlled load services 525 may combine these two schemes (and be quite complicated). In total, if 526 we take into account the need of different charging levels for busy and 527 idle hours, the charging structure tends to be very complicated due to 528 the large amount of parameters. On the contrary, the charging of the 529 SIMA service can be based purely on one parameter, NBR. 531 When a customer requests a service she/he shall inform the network what 532 kind of service is needed. This information consists usually of some 533 traffic parameters (like peak cell rate) and quality parameters 534 (parameters (cell loss ratio, maximum delay), and service class. In 535 order to successfully use a service, the customer shall understand the 536 meaning of these parameters (if they cannot be totally hidden from end- 537 users), and even to make proper guess for the values of every parameter. 538 Taking into account the reluctance of many Internet users to learn 539 technical details, the current service concepts seem to be 540 unsatisfactory in this respect. With SIMA there is only NBR and the 541 selection between real-time and non-real-time service, and moreover, the 542 latter selection can be left usually to the application. 544 The next question is whether a SIMA network can offer all the necessary 545 service types. SIMA can provide efficient real-time service (i.e., as 546 small delay as possible), different packet loss ratios from negligible 547 to high, and a free combination of these two categories (delay, packet 548 loss ratio). The most unclear service class is the controlled load 549 service with small packet loss ratio. However, it should be stressed 550 that a small packet loss ratio can be always attained by using efficient 551 upper layer protocols if there is no strict delay requirement. 552 Therefore, in an environment where most customers are able to use TCP/IP 553 or similar protocol, there is no urgent need for a controlled load 554 service as a service offered to customers, rather the objective of 555 controlled load service is to optimize the use of network resources. A 556 SIMA network offers good possibilities for an application using TCP/IP 557 or similar protocols, as the packet loss ratio always decreases rapidly 558 when the transmission rate goes down enough (say, to a level of 559 2.5*NBR). In this respect SIMA service is essentially better than a pure 560 best-effort service. 562 As a conclusion, SIMA is able to offer simple, feasible solution for all 563 the service needs while traditional approaches call for three quite 564 complex services. Moreover, as the basic idea of SIMA is quite close to 565 the philosophy of the current Internet, SIMA is a natural way to 566 implement new services in Internet. 568 Table 1. Comparison of network services, user related aspects 569 ------------------------------------------------------------------------ 570 | | IntServ | ATM | SIMA | 571 | | guaran. contr. best- | CBR+ ABR UBR | | 572 | | serv. load effort | rt-VBR | | 573 |-----------+----------------------+-----------------------+-----------| 574 | Charging | ? ? flat | based on ? flat | based on | 575 | | rate | traffic, QoS rate | NBR | 576 | | | parameters (+usage)| | 577 | | | | | 578 | Traffic | bucket rate, - | PCR, (about - | | 579 | parameters| bucket size, | SCR, BT 20) | | 580 | | peak rate, | | (NBR) | 581 | | min. policed | | | 582 | | unit, max | | | 583 | | packet size | | | 584 | | | | | 585 |QoS paramet| yes yes no | yes yes no | no | 586 | | | | | 587 |Serv.classe| | | | 588 |small delay| yes no no | yes no no | yes/no | 589 |loss ratio | small small high | small small high | small-high| 590 |controlled | no yes no | no yes no | possible | 591 | service | | | | 592 ------------------------------------------------------------------------ 594 3.3 Management comparison 596 The other important aspect necessary to evaluate is how well the SIMA 597 concept can meet the requirements of network operators (see table 2). 598 The main objective of a network operator is to offer those services that 599 most customers need by a competitive price. Because traffic and network 600 management is one of major costs of telecommunication network, it is 601 very important to keep management functions as simple and efficient as 602 possible. Best-effort service is simple to manage but it does not 603 provide all the needed features, whereas with guaranteed services it i 604 possible to offer all kind of services at the expense of increased 605 complexity. The realization of a guaranteed service requires traffic 606 parameters for every flow, controlling of these parameters, resource 607 reservation at every network node, complicated signaling for the 608 transfer of parameters, capacity planning for every service class, 609 dimensioning of complicated buffer and switching structures, etc. It 610 will be very difficult to implement and manage this type of network. 612 The SIMA service makes possible a very simple management. Simple traffic 613 management means that the operator offers in principle only one service 614 with two components: a real-time service class and a non-real-time 615 service class. Notwithstanding the simplicity, this one basic service is 616 able to offer different quality levels with an automatic charging 617 structure. 619 The technical basis of the SIMA service lays on principles of best- 620 effort service: on the one hand, users do not inform in advance the 621 network on the needed bit rate or any other traffic parameters, and on 622 the other hand, the network operator does not give any precise 623 guarantees of the available bit rates or QoS (Quality of Service). The 624 best-effort principle with the aid of priorities makes possible a simple 625 network structure and management and, at the same time, it results in 626 good fairness among different connections and efficient statistical 627 multiplexing. The basic version of the SIMA service works without such 628 ordinary management functions as Traffic Descriptor, QoS parameters, 629 Service Classes, Connection Admission Control (CAC), or Usage Parameter 630 Control (UPC). All these functions are replaced by two autonomous units: 631 the measuring unit at access nodes and the scheduling and buffering unit 632 (SBU). 634 Table 2. Comparison of network services, network related aspects 635 ------------------------------------------------------------------------ 636 | | IntServ | ATM | SIMA | 637 | | guaran. contr. best- | CBR+ ABR UBR | | 638 | | serv. load effort | rt-VBR | | 639 |----------+-----------------------+-----------------------+-----------| 640 | traffic | yes no | yes yes no | no | 641 | control | | (difficult) | | 642 | per flow | | | | 643 | | | | | 644 |resource | yes no | yes for MBR no | no (or | 645 |allocation| |(difficult) | NBR-based)| 646 | | | | | 647 |buffers | from 3 to thousands | from 3 to thousands | two | 648 |per link | | | | 649 | | | | | 650 |network | complicated | complicated | simple | 651 |management| | | | 652 ------------------------------------------------------------------------ 654 3.4 Performance comparison 656 The SIMA service is able to meet the simplicity requirement essentially 657 better than a network with several service classes, and it can satisfy 658 the basic service needs of most customers. The remaining questions are 659 related to the performance of SIMA networks: is the throughput 660 sufficient without a complicated control system, and could customers 661 rely on the quality of the network without guaranteed services. 663 As to the throughput, the main advantage of SIMA is that there is no 664 need to fragment the network capacity, instead, all services and all 665 flows divide the whole capacity of every link. In this respect SIMA is 666 very efficient. The possible problems are related to the fact that some 667 packet could be lost near the receiver and, therefore, the capacity that 668 these packets have exploited in the previous links will be wasted. The 669 best solution to this problem is the proper network dimensioning. 670 Another problematic issue could be partly transmitted packets, if ATM or 671 similar technology is used as at the transmission level. If this problem 672 turns out be a major problem, a packet based discarding scheme can be 673 implemented throughout the network, but at the expense of keeping track 674 on every flow in every network node. 676 The most difficult and crucial issue with the traffic management of the 677 SIMA service is the dimensioning of the network because it is the best 678 and only available tool to keep customers satisfied with the service. 679 One possible approach is that the operator attempts to offer 680 satisfactory QoS to nominal connections (i.e., to those connections in 681 which actual bit rate is equal to NBR). In practice, this may mean that 682 the operator measures the average cell loss ratio of cells with priority 683 level 4. This ratio should remain on a reasonable level, for instance 684 less than 1E-7. If this cell loss ratio is exceeded continuously, the 685 operator shall firstly identify the bottlenecks in the network and then 686 increase the network capacity in those points. It should be noted that 687 this capacity increase is a quite straightforward task because there is 688 no need to make any new plans concerning switching structure, the 689 capacity division between service classes or virtual paths, etc. The 690 network operator simply throws bandwidth, and the SIMA service manages 691 QoS. This network dimensioning scheme provided by SIMA is a natural 692 extension of the prevalent way of managing Internet. 694 4. Network Element Data Handling Requirements 696 There are two main alternatives for the realization of the SIMA service: 697 the first one based purely on packet network and the second one based on 698 the use of ATM for the switching and transportation. As the basic 699 implementation of these two alternatives does not considerable differ 700 from each other, in the following both versions are presented in 701 parallel. The main difference is that the ATM makes possible to realize 702 more easily a satisfactory real-time service. This question is addressed 703 further in chapter 9.3. 705 The implementation of the SIMA service consists of two main parts: 706 access nodes and core network nodes presented in Fig. 5. There is a 707 fundamental difference between these node types: the traffic measurement 708 of every flow is performed at access nodes whereas at the core network 709 nodes the traffic control functions do not need to know anything about 710 the properties of separate flows. 712 C -------- C -------- 713 / \ 714 +-----+ +----+ / \ +----+ +-----+ 715 | CE1 |---| A1 |--- C --------- C ----------- C -----| A2 | ----| CE2 | 716 +-----+ +----+ \ / / +----+ +-----+ 717 C------ C ------------- 719 Fig. 5. Customer equipment (CE1) connected to an other customer 720 equipment (CE2) through a SIMA network with access nodes (A) and core 721 nodes (C). 723 4.1. Access Node 725 Let us suppose that there is an IP flow (i) at an access node. A nominal 726 bit rate, NBR(i), is associated to the flow and the user is transmitting 727 IP packets (which may be converted into ATM cells) into the network 728 according to an arbitrary traffic process. At the user/network interface 729 there is a measuring device which measures the momentary bit rate of the 730 connection at the arrival of the j:th packet (or cell). This rate is 731 denoted by MBR(i,j). The device gives every packet (or cell) a priority, 732 PL(i,j), based on the MBR(i,j) to NBR(i) ratio: 734 x = 4.5 - ln(MBR(i,j)/NBR(i))/ln(2) 736 PL(i,j) = 6 if x >= 6 737 = Int(x) if 0 < x < 6 (1) 738 = 0 if x <= 0 740 where Int(x) is the integer part of x. 742 Consequently, if MBR(i,j) = NBR(i) the packet (or cell) gets priority 4, 743 if MBR(i,j) > 5.66 NBR(i) the packet (or cell) gets the lowest priority 744 (0), and if MBR(i,j) < 0.17 NBR(i) the packet (or cell) gets the highest 745 NBR-priority (6). Priority 7 is reserved for those connections that use 746 a network service with guaranteed bandwidth and quality. The accepting 747 and discarding of packets (or cells) inside a SIMA network is entirely 748 based on these priorities. 750 This priority scheme makes it possible to offer different QoS levels as 751 regards the packet loss probability. The other needed QoS distinction is 752 related to the delay and delay variation requirements: the network shall 753 be able to guarantee small delay for real-time flows. The user shall 754 determine for every flow whether a real-time service is required. In the 755 SIMA service this selection can be left freely to the user and there is 756 no need to take it into account when determining the charge of the 757 connection. For the realization of real-time service every network node 758 (and every separate switching element in a network node) shall have two 759 parallel buffers: one for real-time flows and another for non-real-time 760 flows. All packets or cells belonging to a real-time flow go through the 761 real-time buffer and all other packets use the non-real-time buffer. 763 4.2 Scheduling and buffering in core network 765 In core network, the key issue in the implementation of the SIMA service 766 is the packet or cell discarding system before the actual buffering 767 shown in Fig. 6. At any instant there is an accepted level of priority 768 (PL_a): if an incoming packet or cell has the same or higher priority, 769 it is accepted, otherwise it is discarded. The calculation of PL_a is 770 based on the buffer occupancy levels of the real-time buffer (M_rt) and 771 non-real-time buffer (M_nrt). 773 All the packets or cells which have been accepted in the scheduling unit 774 are situated either in the real-time or non-real-time buffer (the 775 scheduling algorithm can guarantee that there is no cell loss in actual 776 buffers). Both buffers may apply the First In First Out (FIFO) 777 principle. In order to obtain a small delay and delay variation, the 778 real-time buffer should be relatively small (e.g., 10 Kbytes). All 779 packets (or cells) in the real-time buffer shall be transmitted before 780 any packet (or cell) in the non-real-time buffer. It should be 781 emphasized that the delay priority of real-time flows has no effect on 782 the packet loss ratios. The non-real-time buffer should be much larger 783 (e.g., 1 Mbyte) because of the packet scale fluctuations in typical non- 784 real-time traffic processes. Moreover, large buffers make it possible to 785 offer reasonable service for those flows that are capable of adjusting 786 their bit rate. 788 It should be emphasized that the function of each scheduling and 789 buffering unit (SBU) is independent of all other SBU's; all the tasks of 790 SBU are performed based on the information of incoming packets (or 791 cells), and moreover, all the necessary function for the implementation 792 are described in Fig. 6. Thus, due to the autonomous property of 793 switching units and the unnecessity of resource reservation, the 794 management of the SIMA network is very straightforward. 796 ----------------------- 797 | | 798 | PL_a = F(M_rt,M_nrt)| <- - - - - - - - - - 799 | | <- - - - - - - | 800 ----------------------- | 801 | M_rt packet 802 |PL_a real time | ----+-- or 803 | --------------------> XX|----+-X--> cells 804 packet V | | ------- | out 805 or cells / \ / \ | 806 in / \ yes / \ | | 807 -X---> /PL>= \---->/ rt/ \ | 808 \PL_a / \ nrt / | | 809 \ / \ / | 810 \ / \ / M_nrt | 811 | | nrt -------+--------- | 812 | ----------> XXXXXXXXX|--/--- 813 | no ----------------- open only if 814 | (discard packet or cell) M_rt = 0 815 V 817 Fig. 6. A packet (or cell) scheduling and buffering unit (SBU) for a 818 SIMA network node 820 Only additional information that has to transmitted through the SIMA 821 network is the NBR for each flow. The basic principle of SIMA is that 822 the charge of the flow specifies the obtained NBR and the network has to 823 be able to transmit NBR information between access nodes (from A1 to A2 824 or vice versa in Fig. 5). We may assume that the customer is entitled to 825 divide freely the available NBR between the two directions. With certain 826 services (e.g., radio broadcast) pre-defined NBR can be included in the 827 total service fee. 829 5. Invocation Information 831 At the access node, the SIMA service is invoked by specifying the NBR 832 for the data flow, and by selecting either the real-time or non-real- 833 time service. If a pure flat rate charging scheme is used, the NBR can 834 be assigned to the interface not to any individual flow, as described in 835 chapter 2.1. 837 The value of NBR is measured in bits of IP datagrams (or ATM cells) per 838 second. Value of this parameter may range from 8 to 320 Terabits per 839 second. The selection between real-time or non-real-time has to be 840 always determined. 842 6. Exported Information 844 The SIMA service has no required characterization parameters. Individual 845 implementations may export appropriate implementation-specific 846 measurement and monitoring information. 848 7. Policing 850 There is no actual policing mechanism needed with SIMA service. The 851 policing is replaced by the traffic measurement of each flow and by the 852 priority of every packet or cell. This mechanism forms a similar traffic 853 control as UPC and the CLP (Cell Loss Priority) bit in ATM network, 854 where the network may change the CLP bit if the connection exceeds its 855 sustained cell rate. The measuring and priority determination have been 856 presented in chapter 4.1. 858 8. Ordering and Merging 860 With the SIMA service there is no specific problems with ordering or 861 merging as the priority of each packet or cell determines automatically 862 the treatment of the packet or cell without any additional control 863 function. 865 9. Guidelines for Implementors 867 9.1 Actual bit rate measurement 869 Since the bit rate of every connection may change significantly in 870 several time scales, the network operator must apply an averaging 871 measuring principle to determine the instantaneous cell rate of each 872 connection. The time scale of the measurement shall depend on the 873 service class (real-time or non-real-time) because the non-real-time 874 buffer capacity can be 100 times larger than the real-time one. The 875 approach presented in this chapter is applicable, but any measuring 876 scheme which gives a feasible approximation of the instantaneous bit 877 rate can be used, provided that it can be adjusted to the needed 878 measuring period. 880 This measuring approach is based on the well-known principle of 881 exponential moving average. In this chapter we assume that ATM is used 882 as transport technology. However, a similar measuring principle can be 883 applied with variable size packets as well. 885 If we suppose that the moving average is calculated at every time slot 886 in the access node, the measured load generated by a connection (i) at 887 the instant of transmission of j:th cell is: 889 rho(i,j)=alpha + rho(i,j-1)(1-alpha)^N(i,j) (2) 891 where N(i,j) is the distant between j:th and (j-1):th cells in time 892 slots and alpha is a parameter which defines the time scale of 893 measurement. Here the notation a^b means a to the power of b. Formula 894 (2)is obtained by assuming that the estimation for the instantaneous 895 load is updated at every time slot, but all calculations are performed 896 only at the arrival instant of a cell. The following starting values can 897 be used: rho(i,0) = 0 and N(i,1)=C/NBR_i, where NBR_i is the nominal 898 bit rate assigned to connection i. 900 In order to obtain an exact steady state value for constant bit rate 901 connections the following conversion between load (rho(i,j)) and 902 measured bit rate (MBR(i,j)) shall be applied: 904 MBR(i,j) = C ln(1-alpha)/ln(1-(alpha/rho(i,j))) (3) 906 where C is the link capacity [bit/s] at the user/network interface. For 907 numerical reasons (2) and (3) shall be replaced by 909 MBR(i,j) = C/N(i,j) 911 if N(i,j) > 10/alpha. 913 It should be noted that because alpha is usually a constant, formula (3) 914 can be replaced by a table with a proper granularity. For the same 915 reason, at least the term (1-alpha)^N(i,j) in (2) can be tabulated. 917 The proper value for parameter alpha depends on the buffer capacity 918 reserved for the service class used by the connection. With real-time 919 services (with small delay variation) the buffer shall be small, and 920 thus the value of alpha must be quite high. On the contrary, when using 921 a non-real-time service the user may want to send bursts of cells 922 without high cell loss ratio. As a consequence alpha must be much 923 smaller (or the averaging period should be much longer). As an interim 924 approach the following approximation might be applicable: 926 alpha = 5/K_n (4) 928 where K_n is the buffer capacity in cells reserved for the service class 929 n. 931 9.2 Implementation of scheduling algorithm 933 The key point of the SIMA service lies in the function of the scheduling 934 algorithm. The decision of the acceptance is based on two parameters: 935 the priority level of the cell and the occupancy level of the two 936 buffers. Let us use the following notations: 938 M_rt = the number of cells in the rt-buffer 939 K_rt = the number of buffer places in the rt-buffer 940 M_nrt = the number of cells in the nrt-buffer 941 K_nrt = the number of buffer places in the nrt-buffer 943 The average occupancy level of the total buffering system (x) might be 944 determined in several ways, for instance: 946 x = (x_rt + x_nrt) (a) 947 x = sqrt(x_rt^2 + x_nrt^2) (b) (5) 948 x = max(x_rt, x_nrt) (c) 949 where: 950 x_rt = M_rt/K_rt 951 x_nrt = M_nrt/K_nrt 953 Above sqrt(y) stands for taking squareroot from y and max(y,z) stands 954 for taking the maximum of y and z. 956 The cell is accepted if the following relation is valid 958 PL < a - b*x (6) 960 In reality formulae (5) and (6) can be implemented by using pre- 961 calculated tables. 963 9.3 ATM implementation 965 One of the main question related with the implementation is whether ATM 966 is used as the transmission technology. The main advantage of using ATM 967 in a SIMA network is that it makes possible a real-time service with 968 small delay variation. There are some drawbacks if ATM is not used. 970 Firstly, if a pure packet based approach is applied, the delay variation 971 of real-time connections will be much larger because of transmission of 972 long non-real-time packets. One solution to this problem could be to 973 interrupt the transmission of non-real-time packet if needed (something 974 similar to ATM). This problem is significant especially with low link 975 speeds. 977 Secondly, the function of the network shall be realized in a way that 978 either there cannot be long non-real-time packets, or long real-time 979 packets must always be assigned with the lowest priority. The reason to 980 this requirement is that even a couple of long non-real-time packets 981 could increase the buffer occupancy level of real-time buffer, and by 982 that means change very rapidly the allowed priority level. 984 9.4 The location of priority bits 986 SIMA needs at most 4 bits in every packet in pure IP networks or in 987 every cell if ATM is used. In IP packets it may be possible to use the 988 type of service (TOS) field or priority field if IP-version 6 is used. 990 In ATM the cell header is small and there is only one bit for priority 991 marking (CLP bit). There are at least three different approaches to 992 situate the needed bits in ATM cells. 994 Firstly, it could be possible to use the Generic Flow Control (GFC) 995 field in the beginning of the cell. It should be noted that this field 996 is actually available only in UNI-interface, as in the NNI-interface the 997 first 4 bits belongs to the VPI-field. However, the first 4 bits are not 998 widely used in real implementations. 1000 Secondly, it could be possible to reserve an own VP for each of the 1001 possible 16 service types: 8 priority levels for real-time connections 1002 and 8 priority levels for non-real-time connections. The advantage of 1003 this scheme is that it has a good interoperability with current 1004 networks, as it does not change the cell structure at all. The main 1005 disadvantage is that it reduces the number of available VP for other 1006 purposes. However, if SIMA is widely applied there is not so much need 1007 for using separate VPs for different types of services, rather the VPs 1008 can be used only for routing purposes. 1010 Thirdly, if it is not possible to use the header of the ATM cells, the 1011 bits should be situated in the information field. However, this is not a 1012 desirable approach, because it is very difficult to find any place 1013 suitable to all ATM adaptation layers (AAL). 1015 10. Evaluation Criteria 1017 The evaluation of SIMA service consists of three main parts: functions 1018 related to priority determination, scheduling functions, and the 1019 performance of real-time service. 1021 The assumption of this document is that the measuring principle will not 1022 be standardized, although one possible implementation has been presented 1023 in chapter 9.1. However, it can assumed that all measuring 1024 implementations shall produce the same result for constant bit rate 1025 (CBR) connections with certain accuracy. That means that for certain 1026 constant bit rate, the priority shall be within certain limits. If this 1027 priority determination is accurate, it makes possible for the user to 1028 optimize the bit rate he/she is using. Therefore, the main requirement 1029 for an realization of the priority determination of CBR stream is that 1030 it should not give worse priority than what the formula (1) gives. 1031 Because the SIMA service is to certain extent indefinite, there is no 1032 need to define any stringent limit for the underestimation of the actual 1033 bit rate, but, for instance, 1 percent accuracy can be sufficient. 1035 The implementation of scheduling function does not need to be 1036 standardized (some examples are presented in chapter 9.2). The operator 1037 is allowed to optimize the network performance by changing the 1038 scheduling function. 1040 All real-time connections or flows shall have as small delay variation 1041 as possible. However, as the actual delay variation depends essentially 1042 on the underlying technology used, it is not feasible to define any 1043 numerical requirements for the delay variation (see chapter 9.3). The 1044 only requirement is that if there is any real-time packet or cell in the 1045 buffer, it shall be transmitted before all the packets or cells in the 1046 non-real-time buffer. 1048 11. Examples of Implementation 1050 The basic implementation approach for an Internet Service providers is 1051 based on the flat rate charging scheme. This implementation of Internet 1052 service is very straightforward expansion of the current Internet 1053 service. 1055 Every customer has certain permanent NBR which has direct relation to 1056 the monthly charge of the customer. All the traffic sent by the customer 1057 to the network is measured at access node of the network. If the access 1058 is based on IP, the node gives every packet a priority based on the 1059 measurement result and the NBR of the customer, as presented in chapter 1060 4.1. Correspondingly, if the access is based on ATM, the node gives 1061 priority for every cell. Priorities from 0 to 6 are used for SIMA 1062 customer. Priority 7 can be used by customer which needs reserved 1063 connections, and which are ready to pay more than the SIMA customers. 1064 The lowest charging category with minimal NBR could be very inexpensive, 1065 as those customers always get priority 0, and therefore, cannot disturb 1066 the traffic of other customers. 1068 12. Examples of Use 1070 The main advantage of SIMA as an Internet service is that it provides a 1071 uniform service concept for different needs from file transfer 1072 applications using TCP/IP protocol without loose delay and packet loss 1073 requirements to real-time applications with very strict quality and 1074 availability requirements. Thus SIMA can be used as a complete Internet 1075 service for all customers. 1077 13. Security Considerations 1079 Security considerations are not discussed in this memo. 1081 Author's address 1082 Kalevi Kilkki 1083 Nokia Research Center 1084 P.O.Box 422 1085 FIN-00045 NOKIA GROUP 1086 Finland 1087 E-mail: kalevi.kilkki@research.nokia.com 1088 Tel. + 358 9 4376 6817 1089 Fax. + 358 9 4376 6851 1090 Information about SIMA is available from http://www-nrc.nokia.com/sima/ 1091 Expiration 1092 This document will expire in 19 December 1997.