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 ...
maximum throughput possible is 0.78 (mean ABR capacity). The e ciency is cal- culated as follows. We rst measure the the aggregate mean VBR rate (since it is not the sum of the individual mean rates due to bounding the values to 0 and 15 Mbps). Subtract it from 149.76 Mbps to get the mean ABR capacity. Then multiply the ABR capacity by 0.78 (or 0.87) to get the maximum possible throughput. We then take the ratio of the measured TCP throughput and this calculated value to give the e ciency. 8.18.1 E ect of High Variance and Total VBR Load In this section, we present simulation results where we vary the mean and the standard deviation of the individual video sources such that the total variance is always high, and the total maximum VBR load varies. In Table 8.10, and Table 8.11, we show the maximum queue length, the total TCP throughput, VBR throughput, ABR throughput, and e ciency for three combinations of the mean and standard deviation. Table 8.11 is for TCP MSS = 512 bytes, while Table 8.11 is for TCP MSS = 9140 bytes. Video Sources ABR Metrics # Mean Standard Max Switch Q Total E ciency per-source Deviation (cells) TCP (%of Max rate (Mbps) (Mbps) T'put throughput) 1. 5 5 6775 (1.8 F/b Delay) 68.72 Mbps 94.4% 2. 7.5 7 7078 (1.9 F/b Delay) 59.62 Mbps 94.1% 3. 10 5 5526 (1.5 F/b Delay) 82.88 Mbps 88.4% Table 8.10: E ect of Variance and VBR Load: MSS = 512 bytes 323
Video Sources ABR Metrics # Mean Standard Max Switch Q Total TCP E ciency per-source Deviation (cells) Throughput (%of Max rate (Mbps) (Mbps) throughput) 1. 5 5 5572 (1.5 F/b Delay) 77.62 Mbps 95.6% 2. 7.5 7 5512 (1.5 F/b Delay) 67.14 Mbps 95% 3. 10 5 5545 (1.5 F/b Delay) 56.15 Mbps 95.6% Table 8.11: E ect of Variance and VBR Load: MSS = 512 bytes Observe that the measured mean VBR thoughput (column 6) is the same in corresponding rows of both the tables. This is because irrespective of ABR load, VBR load is given priority and cleared out rst. Further, by bounding the MPEG-2 SPTS source rate values between 0 and 15 Mbps, we ensure that the total VBR load is about 80of the link capacity. For row 1, measured VBR throughput (column 6) was 56.44 Mbps (against 9 5 = 45 Mbps expected without bounding). For row 2, it was 68.51 Mbps (against 9 7.5 = 67.5 Mbps expected without bounding). For row 3, it was 82.28 Mbps(against 9 10=90Mbpsexpected without bounding). Observe that when the input mean is higher, the expected aggregate value is lower and vice-versa. The e ciency values are calculated using these values of total VBR capacity. For example, in row 1 of Table 8.10, the ABR throughput is is 149.76 - 56.44 = 93.32 Mbps. ForaMSSof 512, the maximum TCP thoughput is 78% of ABR throughput = 72.78 Mbps (not shown in the table). Given that TCP thoughput achieved is 68.72 Mbps (Column 5), the e ciency is 68.72/72.78 = 94.4%. For Table 8.11, since the 324
- 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
- Page 333 and 334: hand, the frequency of the VBR is h
- Page 335 and 336: like ERICA+ which uses the queueing
- Page 337 and 338: minimum fairshare is low. This may
- Page 339 and 340: 8.17 E ect of Long-Range Dependent
- Page 341 and 342: terminology) are called \Presentati
- Page 343 and 344: The key point is that the MPEG-2 ra
- Page 345 and 346: the MPEG-2 Transport Stream, and st
- Page 347 and 348: 8.17.4 Observations on the Long-Ran
- Page 349: For the video sources, we choose me
- Page 353 and 354: However, with modi cations to ERICA
- Page 355 and 356: Video Sources ABR Metrics # Avg Src
- Page 357 and 358: Video Sources ABR Metrics # Avg Src
- Page 359 and 360: On the other hand, if the applicati
- Page 361 and 362: of managing bu ers, queueing, sched
- Page 363 and 364: hence control the total load on the
- Page 365 and 366: call such a switch a \VS/VD switch"
- Page 367 and 368: Figure 9.2: Per-class queues in a n
- Page 369 and 370: which arises is where the rate calc
- Page 371 and 372: 9.2 The ERICA Switch Scheme: Renota
- Page 373 and 374: The unknowns in the above equations
- Page 375 and 376: Figure 9.9: Two methods to measure
- Page 377 and 378: 9.4 VS/VD Switch Design Options 9.4
- Page 379 and 380: # VC Rate VC Input Rate Input Rate
- Page 381 and 382: 8 uses source rate measurement, we
- Page 383 and 384: The allocated rate update and the e
- Page 385 and 386: sources in chapter 6. We expect the
- Page 387 and 388: con guration mentioned in the table
- Page 389 and 390: can be very di erent for di erent V
- Page 391 and 392: CHAPTER 10 IMPLEMENTATION ISSUES At
- Page 393 and 394: With an enhanced UBR service, appli
- Page 395 and 396: 2. Some switch schemes have a proce
- Page 397 and 398: 4. Large legacy switches have a pro
- Page 399 and 400: section 9. Further, WAN switches wo
maximum throughput possible is 0.78 (mean <strong>ABR</strong> capacity). The e ciency is cal-<br />
culated as follows. We rst measure <strong>the</strong> <strong>the</strong> aggregate mean VBR rate (s<strong>in</strong>ce it is not<br />
<strong>the</strong> sum of <strong>the</strong> <strong>in</strong>dividual mean rates due to bound<strong>in</strong>g <strong>the</strong> values to 0 and 15 Mbps).<br />
Subtract it from 149.76 Mbps to get <strong>the</strong> mean <strong>ABR</strong> capacity. Then multiply <strong>the</strong><br />
<strong>ABR</strong> capacity by 0.78 (or 0.87) to get <strong>the</strong> maximum possible throughput. We <strong>the</strong>n<br />
take <strong>the</strong> ratio of <strong>the</strong> measured TCP throughput and this calculated value to give <strong>the</strong><br />
e ciency.<br />
8.18.1 E ect of High Variance and Total VBR Load<br />
In this section, we present simulation results where we vary <strong>the</strong> mean and <strong>the</strong><br />
standard deviation of <strong>the</strong> <strong>in</strong>dividual video sources such that <strong>the</strong> total variance is<br />
always high, and <strong>the</strong> total maximum VBR load varies.<br />
In Table 8.10, and Table 8.11, we show <strong>the</strong> maximum queue length, <strong>the</strong> total TCP<br />
throughput, VBR throughput, <strong>ABR</strong> throughput, and e ciency <strong>for</strong> three comb<strong>in</strong>ations<br />
of <strong>the</strong> mean and standard deviation. Table 8.11 is <strong>for</strong> TCP MSS = 512 bytes, while<br />
Table 8.11 is <strong>for</strong> TCP MSS = 9140 bytes.<br />
Video Sources <strong>ABR</strong> Metrics<br />
# Mean Standard Max Switch Q Total E ciency<br />
per-source Deviation (cells) TCP (%of Max<br />
rate (Mbps) (Mbps) T'put throughput)<br />
1. 5 5 6775 (1.8 F/b Delay) 68.72 Mbps 94.4%<br />
2. 7.5 7 7078 (1.9 F/b Delay) 59.62 Mbps 94.1%<br />
3. 10 5 5526 (1.5 F/b Delay) 82.88 Mbps 88.4%<br />
Table 8.10: E ect of Variance and VBR Load: MSS = 512 bytes<br />
323