Traffic Management for the Available Bit Rate (ABR) Service in ...
Traffic Management for the Available Bit Rate (ABR) Service in ... Traffic Management for the Available Bit Rate (ABR) Service in ...
may reschedule upon receipt of new feedback if the next cell transmission opportunity calculated with the new ACR is earlier than the one currently scheduled. In terms of pseudo code: IF( time to send > (now + 1/acr) THEN time to send (now + 1/acr) This mechanism is also illustrated in gure 7.14. 7.3 Summary of Source Rule Design Issues As explained in chapter 2, source and destination end system rules are important in complimenting the switch feedback calculation mechanisms. Speci cally, the source rules provide \open-loop" control which is e ective in cases when the source starts sending data after idle periods, and/or when the switch feedback to the sources is disrupted. Further, the sources have to consider the scheduling of RM and data cells, especially in the case of low rate tra c. This dissertation work has addressed the standardization aspects of Use-it-or-Lose-It policies (the issue arises when sources start sending data after idle periods), and that of low rate sources. Speci cally, this work has helped design some of the source rules of the international standard (SES Rules 5, 9, 11, and 13). 255
CHAPTER 8 SUPPORTING INTERNET APPLICATIONS OVER THE ATM-ABR SERVICE With the proliference of multimedia tra c over the Internet, it seems natural to move over to ATM technology which has been designed speci cally to support integration of data, voice, and video applications. While multimedia applications are still in the development stage, most of the tra c on the Internet today is data tra c in the sense that they are bursty and relatively delay insensitive. It is, therefore, natural to ask how the current applications will perform over the ATM technology. Although ATM technology has been designed to provide an end-to-end transport level service and so, strictly speaking, there is no need to have TCP or IP if the entire path from source to destination is an ATM path. However, in the forseeable future, this scenario is going to be rare. A more common scenario would be where only part of the path is ATM. In this case, TCP is needed to provide the end-to-end transport functions (like ow control, retransmission, ordered delivery) and ATM networks are used simply as "bit pipes" or "bitways." Since the Available Bit Rate (ABR) and the Unspeci ed Bit Rate (UBR) service classes have been developed speci cally to support data applications, it is important to investigate the performance of dominant internet applications like le transfer and 256
- Page 231 and 232: (a) Transmitted Cell Rate (c) Link
- Page 233 and 234: (a) Transmitted Cell Rate (basic ER
- Page 235 and 236: (a) Transmitted Cell Rate (basic ER
- Page 237 and 238: (a) Transmitted Cell Rate (c) Link
- Page 239 and 240: (a) Transmitted Cell Rate (c) Link
- Page 241 and 242: (a) Transmitted Cell Rate (c) Link
- Page 243 and 244: (a) Transmitted Cell Rate (c) Link
- Page 245 and 246: (a) Transmitted Cell Rate (c) Link
- Page 247 and 248: (a) Transmitted Cell Rate (c) Link
- Page 249 and 250: (a) Transmitted Cell Rate (basic ER
- Page 251 and 252: RM cells. In this chapter, we devot
- Page 253 and 254: low. This tradeo was discovered and
- Page 255 and 256: in the speci cation. b) ACR shall n
- Page 257 and 258: 7.1.6 December 1995 Proposals There
- Page 259 and 260: Figure 7.2: Multiplicative vsAdditi
- Page 261 and 262: is never triggered. However, the PR
- Page 263 and 264: simulation results of bursty source
- Page 265 and 266: 1. The time-based proposal also ind
- Page 267 and 268: The switch maintains a local alloca
- Page 269 and 270: PNI = f0, 1g : f1 ) No rule 5b, 0 )
- Page 271 and 272: after reaching the goal. The time-b
- Page 273 and 274: Figure 7.8: Closed-Loop Bursty Tra
- Page 275 and 276: The algorithm measures the load and
- Page 277 and 278: Medium Bursts Medium bursts are exp
- Page 279 and 280: count-based technique may be insu c
- Page 281: Figure 7.13: An event trace illustr
- Page 285 and 286: u ering which does not depend upon
- Page 287 and 288: 4 make it to the destination are ar
- Page 289 and 290: Figure 8.3: At the ATM layer, the T
- Page 291 and 292: issues and e ect of bursty applicat
- Page 293 and 294: from the loss and they trigger the
- Page 295 and 296: 8.8 Performance Metrics We measure
- Page 297 and 298: the performance is fair. Also, the
- Page 299 and 300: Figure 8.7(b) shows the rates (ACRs
- Page 301 and 302: Window Size in bytes vfive-tcp/opti
- Page 303 and 304: The e ect of large bu ers on CLR is
- Page 305 and 306: 8.13 Summary of TCP/IP performance
- Page 307 and 308: Feedback delay: Twice the delay fro
- Page 309 and 310: on a link, two cells are expected a
- Page 311 and 312: state only after the switch algorit
- Page 313 and 314: queue length is less susceptible to
- Page 315 and 316: to equalize rates for fairness, and
- Page 317 and 318: QB = Link bandwidth (RT T ; T ) and
- Page 319 and 320: u ered at the end-system, and not i
- Page 321 and 322: Part b): When ABR load goes away, t
- Page 323 and 324: All our simulations presented use t
- Page 325 and 326: Averaging RTT(ms) Feedback Max Q Th
- Page 327 and 328: a modi ed version of the ERICA algo
- Page 329 and 330: ABR is better than UBR in these (en
- Page 331 and 332: problems. During the ON time, the V
may reschedule upon receipt of new feedback if <strong>the</strong> next cell transmission opportunity<br />
calculated with <strong>the</strong> new ACR is earlier than <strong>the</strong> one currently scheduled. In terms of<br />
pseudo code:<br />
IF( time to send > (now + 1/acr)<br />
THEN time to send (now + 1/acr)<br />
This mechanism is also illustrated <strong>in</strong> gure 7.14.<br />
7.3 Summary of Source Rule Design Issues<br />
As expla<strong>in</strong>ed <strong>in</strong> chapter 2, source and dest<strong>in</strong>ation end system rules are important<br />
<strong>in</strong> compliment<strong>in</strong>g <strong>the</strong> switch feedback calculation mechanisms. Speci cally, <strong>the</strong> source<br />
rules provide \open-loop" control which is e ective <strong>in</strong> cases when <strong>the</strong> source starts<br />
send<strong>in</strong>g data after idle periods, and/or when <strong>the</strong> switch feedback to <strong>the</strong> sources is<br />
disrupted. Fur<strong>the</strong>r, <strong>the</strong> sources have to consider <strong>the</strong> schedul<strong>in</strong>g of RM and data cells,<br />
especially <strong>in</strong> <strong>the</strong> case of low rate tra c. This dissertation work has addressed <strong>the</strong><br />
standardization aspects of Use-it-or-Lose-It policies (<strong>the</strong> issue arises when sources<br />
start send<strong>in</strong>g data after idle periods), and that of low rate sources. Speci cally, this<br />
work has helped design some of <strong>the</strong> source rules of <strong>the</strong> <strong>in</strong>ternational standard (SES<br />
Rules 5, 9, 11, and 13).<br />
255