ecn or delay or both - people | mit csail · 2019-05-29 · yibozhu, monia ghobadi, vishal misra...
TRANSCRIPT
![Page 1: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/1.jpg)
VishalMisraColumbiaUniversity(andGoogle*)
PresentingworkofMonia Ghobadi,Jitu Padhye,KKRamakrishnan,MohitTahiliani andYibo Zhu
CongestionControl Protocols: ECNorDelayorBoth?
![Page 2: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/2.jpg)
Yibo Zhu,Monia Ghobadi, VishalMisra andJitendra Padhye, ECNorDelay:Lessons LearntfromAnalysis ofDCQCNandTIMELY,Proceedings of2016ACMConference onEmergingnetworkexperiment andtechnology (CoNEXT 2016),December, 2016.
![Page 3: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/3.jpg)
• fairnessgoal:
• scalabilitygoal:
TCP connection 1
bottleneckrouter/switchcapacity RTCP connection 2
![Page 4: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/4.jpg)
R
RConnection 1 sending rate
Networkcongested
Networkuncongested
congested not
increaseuncongested
decrease congested
![Page 5: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/5.jpg)
R
RConnection 1 throughput
No congestion: additive increaseCongestion: decrease window by factor of 2No congestion: additive increaseCongestion: decrease window by factor of 2
Network congested
Network uncongested
![Page 6: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/6.jpg)
•
•
•
•
•
•
W
W/2
![Page 7: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/7.jpg)
AQMSender Receiver
PacketMark
AcksReflectMark
![Page 8: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/8.jpg)
![Page 9: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/9.jpg)
AnalyzingDelayvsECNforRDMA
![Page 10: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/10.jpg)
DCQCN(Microsoft)
ECN
TIMELY(Google)
Delay
DCQCNandTIMELY:CongestionControlforROCEv2
![Page 11: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/11.jpg)
Wedon’t haveanintuitive explanation
![Page 12: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/12.jpg)
![Page 13: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/13.jpg)
![Page 14: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/14.jpg)
![Page 15: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/15.jpg)
• Feedbackisdelayedasqueuebuildsup
![Page 16: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/16.jpg)
T0,Q=2
T1,Q=3
T2,Q=4
Bluepacketarrivalcomplete
Bluepacketisabouttoarrive
Bluepacketreadytodepart…andismarked,reflectingstateofqueueatT2
Marking threshold =4packets
![Page 17: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/17.jpg)
T0,Q=2
T1,Q=3
T2,Q=4
Bluepacketarrivalcomplete.…timerstarts
Bluepacketisabouttoarrive
Bluepacketreadytodepart…andreflectsstateofqueueatT0
![Page 18: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/18.jpg)
• Delayinherentlyreports“stale”information
• Thestalenessisaffectedbyqueuelength!
• Longerqueueè morestalefeedback
• Thiscanleadtoinstability
![Page 19: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/19.jpg)
• Canhavefixedqueueorfairness– butnotboth!
![Page 20: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/20.jpg)
Bottleneckqueueisafunctionofnumberofflows.
![Page 21: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/21.jpg)
*C.V.Hollot,VishalMisra,DonTowsley andWei-BoGong,OnDesigningImprovedControllersforAQMRoutersSupportingTCPFlows,ProceedingsofIEEEInfocom,April,2001.
![Page 22: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/22.jpg)
0 5
10 15 20 25
0 0.5 1 1.5 2
Rate
(Gbp
s)
Time(s)
4 µs feedback delay
2 flows10 flows64 flows
1
10
100
1000
0 0.5 1 1.5 2
Queu
e(KB
)
Time(s)
4 µs feedback delay
2 flows10 flows64 flows
![Page 23: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/23.jpg)
0
2
4
6
8
10
0 1 2 3 4 5
Rate
(Gbp
s)
Time(s)
1
10
100
1000
0 1 2 3 4 5
Queu
e(KB
)
Time(s)
1
10
100
1000
0 1 2 3 4 5
Queu
e(KB
)
Time(s)
0
0.5
1
1.5
2
0 1 2 3 4 5
Rate
(Gbp
s)Time(s)
![Page 24: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/24.jpg)
![Page 25: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/25.jpg)
• Canhavefixedqueueorfairness– butnotboth!
• ECNmarkingisresistanttofeedback jitter
![Page 26: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/26.jpg)
0 20 40 60 80
100 120 140
0 0.05 0.1 0.15 0.2
Que
ue(K
B)
Time(s)
TIMELYDCQCN
distorted
Delay:ECN::AM:FM
![Page 27: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/27.jpg)
Delay
ECNRoCEv2
+ Generally stable+ Fair & converging - Affected by bufferbloat - Delay increases with the number of flows - Sensitive to variable feedback delay
+ Generally stable - Unfair+ Addressed bufferbloat+ Fixed delay regardless of the number of flows - Sensitive to variable feedback delay
(Patched) TIMELY
PI
PI
+ Generally stable+ Fair & converging+ Addressed bufferbloat - Delay increases with the number of flows+ Resilient to variable feedback delayDCQCN
+ Generally stable+ Fair & converging+ Addressed bufferbloat+ Fixed delay regardless of the number of flows+ Resilient to variable feedback delay
![Page 28: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/28.jpg)
BacktoDelayvsECNforTCP
![Page 29: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/29.jpg)
RConnection 1 sending rate
R
![Page 30: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/30.jpg)
RConnection 1 sending rate
R
![Page 31: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/31.jpg)
![Page 32: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/32.jpg)
![Page 33: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/33.jpg)
![Page 34: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/34.jpg)
![Page 35: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/35.jpg)
PI + ECN BBR DCTCP
Gbps % Gbps % Gbps %
Triumph 1 10.00 100 4.09 40.9 10.00 100Scorpion 10.00 100 4.09 40.9 10.00 100
Triumph 2 1.00 100 0.66 66.0 1.00 100
SwitchesPI + ECN BBR DCTCP
Triumph 1 270095 marks 0 6480390 marks
Scorpion 0 marks 0 0 marks
Triumph 2 192474 marks 0 776279 marks
![Page 36: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/36.jpg)
![Page 37: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/37.jpg)
![Page 38: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/38.jpg)
![Page 39: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/39.jpg)
![Page 40: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/40.jpg)
![Page 41: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/41.jpg)
![Page 42: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/42.jpg)
PI + ECN BBRMbps % Mbps %
L1 50.00 100 49.14 98.28L2 100.00 100 99.35 99.35L3 50.00 100 49.31 98.62L4 150.00 100 148.24 98.83L5 150.00 100 148.76 99.17L6 50.00 100 49.31 98.62
NewReno + PI + ECN BBR
R1 8 marks 0
R2 57 marks 0
R3 42 marks 636 drops
R4 96 marks 1623 drops
R5 182 marks 0
R6 343 marks 3417 drops
![Page 43: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/43.jpg)
![Page 44: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/44.jpg)
![Page 45: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/45.jpg)
ECNAURDelayCanwecombinethetwo?
![Page 46: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/46.jpg)
AQMSender Receiver
PacketMark
Sender Receiver
SendersideAQMPacketMark
![Page 47: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/47.jpg)
allfixedpointsfrequentlyfair
R
R
equal bandwidth share
Connection 1 throughput
![Page 48: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/48.jpg)
Figure 11: Impact of multiple bottleneck links.
Figure 12: Response to sudden changes in respon-sive traffic.
ments investigate the responsiveness of the different schemesto sudden changes in traffic.
In this experiment, 25 PERT flows are started at time 0seconds. Starting at 100 seconds, for the next 300 seconds 25new flows are added at 100 second intervals, causing severecontention for available bandwidth. Starting at 400 seconds,25 flows leave the network at 100 second intervals creatinga sudden availability of bandwidth. We repeat the exper-iment with SACK/Droptail, SACK/RED-ECN and Vegas.Figure 12 shows the aggregate throughput of the set of flowsthat start together. From the figure it is clear that thePERT flows respond quickly to dynamic changes in networkbandwidth. Vegas exhibits previously observed unfairnessamong competing flows. The results for SACK/Droptailand SACK/RED-ECN are similar to that of PERT and areavailable in [4].
We have conducted additional experiments, where dy-namic changes in traffic were caused by non-responsive traf-fic. The results are similar to those above. We have notincluded the results here due to the lack of space. They areavailable in [4].
5. MODELING AND STABILITY OF PERT
5.1 ModelOur modeling of PERT is composed of three parts: win-
dow adjustment, RED emulation, and queuing behavior. Westart with the window dynamics. Similar to [23], we considera single-link scenario and assume the forward propagation
delay from the source to the router is negligible and thus theround-trip time R(t) measured by the end-user at time t iscomposed of backward propagation delay Tp and queuingdelay Tq(t − R(t)), i.e.,
R(t) = Tp + Tq
`
t − R(t)´
. (2)
Denoting by C the link’s capacity and by q(t) the queuesize at time t, queuing delay Tq(t) can be approximated byq(t − R(t))/C. Note that delay R(t) in the last expressionis because the queuing delay perceived by the user at time tis actually experienced by the router R(t) time units earlier.To compare the stability of PERT to router-based RED withstandard TCP, we set window decrease factor β to 0.5 andnote that results for β = 0.35 can be similarly obtainedfollowing the procedure below. Then, window dynamics ofa PERT end-flow is written as:
W (t) =1
R(t)−
W (t)W`
t − R(t)´
2R`
t − R(t)´ p(t), (3)
where, at time t, W (t) is the congestion window size, R(t) isthe RTT, and p(t) is the packet dropping probability. Notethat loss rate p(t) in the last equation is an instantaneousvalue as opposed to its delayed counterpart p(t−R(t)) in theTCP/RED model obtained in [23]. This is because a PERTuser makes its dropping decision at the end-host instead ofthe router.
To formulate PERT’s emulation of the RED mechanism,assume that the propagation delay Tp is known to the end-flow (this can be approximated by the base RTT). Then,upon each packet arrival, the user can estimate the queuingdelay by Tq(t) = R(t) − Tp and generate the packet dropprobability p(t) as following:
p(t) =Tq(t) − Tmin
Tmax − Tminpmax, (4)
where Tmin and Tmax are the maximum and minimum thresh-olds of queuing delays and pmax is a constant.
Another component of RED emulation is the estimationof round-trip time R(t), which is updated per-packet usinga low-pass filter (LPF) with weight α, i.e.,
R(t) = αR(t − 1) + (1 − α)R(t), (5)
where R(t) is the instantaneous RTT measured at time tand weight α = 0.99. Following the technique used in [23],this LPF can be approximated by the following differentialequation:
R(t) =ln αδ
(R(t) − R(t)), (6)
where δ is the sampling interval.We next model the queuing dynamics, which can be de-
scribed by the following differential equation of queue size:
q(t) =W (t)R(t)
N(t) − C,
where N(t) is the number of flows accessing the router attime t and term W (t)N(t)/R(t) can be interpreted as thecombined incoming rate y(t). Since Tq(t) = q(t − R(t))/C,we re-write the last equation in terms of queuing delay Tq(t):
Tq(t) =W (t − R(t))N(t − R(t))
R(t − R(t))C− 1. (7)
8
EmulatingAQMfromEndHosts
![Page 49: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/49.jpg)
![Page 50: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/50.jpg)
uniquefixedpoint
AIMDwithfixedparameters fairfixedpointand fixeddelay
BBRstylecongestioncontrol differentoperatingpoints
![Page 51: ECN or Delay or both - People | MIT CSAIL · 2019-05-29 · YiboZhu, Monia Ghobadi, Vishal Misra and Jitendra Padhye, ECN or Delay: Lessons Learnt from Analysis of DCQCN and TIMELY,](https://reader035.vdocument.in/reader035/viewer/2022070904/5f71972fc6257528aa565a52/html5/thumbnails/51.jpg)
all
R
RConnection 1 sending rate
Networkcongested
Networkuncongested