1. The War Between Mice and Elephants By Liang Guo (Graduate Student) Ibrahim Matta (Professor) Boston University ICNP’2001 Presented By Preeti Phadnis

2. Outline Introduction Introduction Analyzing Short TCP Flow Performance Analyzing Short TCP Flow Performance Architecture and Mechanism –RIO-PS Architecture and Mechanism –RIO-PS Simulations Simulations Discussions Discussions Conclusions and Future work Conclusions and Future work

3. Mice vs Elephants Mice : Short TCP flows e.g. Web Traffic Mice : Short TCP flows e.g. Web Traffic 20% of internet traffic is carried by large number of mice 20% of internet traffic is carried by large number of mice Elephants : Long TCP flows e.g. FTP Elephants : Long TCP flows e.g. FTP 80% of internet traffic is carried by small number of elephants 80% of internet traffic is carried by small number of elephants

4. Internet today WWW : ” World Wide Wait “ term coined by R. Khare and I.Jacobs WWW : ” World Wide Wait “ term coined by R. Khare and I.Jacobs Users spend long time downloading a plain text webpage Users spend long time downloading a plain text webpage Reason: The mice don’t get the fair share of the network resources Reason: The mice don’t get the fair share of the network resources

5. Factors effecting the performance of mice TCP tries to conservatively ramp up its transmission rate to the maximum available bandwidth TCP tries to conservatively ramp up its transmission rate to the maximum available bandwidth For short connections, since congestion window is very small, packet loss always requires timeout to detect. For short connections, since congestion window is very small, packet loss always requires timeout to detect. For the first few packets, since no sampling data is available, TCP has to use a conservatively estimated ITO value as RTO. Short Connection performance is degraded due to large timeout period. For the first few packets, since no sampling data is available, TCP has to use a conservatively estimated ITO value as RTO. Short Connection performance is degraded due to large timeout period.

6. Related work Crovella et al 2001 [16] and Bansal et al 2001 [17] comment that size aware job scheduling helps enhance the response time of short jobs without hurting the performance of long jobs. Crovella et al 2001 [16] and Bansal et al 2001 [17] comment that size aware job scheduling helps enhance the response time of short jobs without hurting the performance of long jobs. D.D Clark and W.Fang 1998 [4] AQM scheme using RED with In and Out Policy D.D Clark and W.Fang 1998 [4] AQM scheme using RED with In and Out Policy

7. Outline Introduction Introduction Analyzing Short TCP Flow Performance Analyzing Short TCP Flow Performance Architecture and Mechanism –RIO-PS Architecture and Mechanism –RIO-PS Simulations Simulations Discussions Discussions Conclusions and Future work Conclusions and Future work

8. Sensitivity Analysis for Short and Long TCP Flows

9. Sensitivity Analysis of Transmission Time

10. Factors Effecting Variability When Loss rate high TCP Congestion control is more likely to enter exponential back off phase, which can cause significantly high variability in transmission time of each individual packet of a flow. Short flows are effected more due to this reason. When Loss rate high TCP Congestion control is more likely to enter exponential back off phase, which can cause significantly high variability in transmission time of each individual packet of a flow. Short flows are effected more due to this reason. When loss rate low, TCP either in slow start or congestion avoidance phase. This dimension of variability is more pronounced for long flows. When loss rate low, TCP either in slow start or congestion avoidance phase. This dimension of variability is more pronounced for long flows.

11. Preferential Treatment to Short TCP flows Simulation using NS simulator Simulation using NS simulator 10 long(10000-packet) TCP-NewReno flows and 10 short(100-packet) TCP- Newreno flows over 1.25Mbps link. 10 long(10000-packet) TCP-NewReno flows and 10 short(100-packet) TCP- Newreno flows over 1.25Mbps link. Queue Management Policy – Drop Tail, RED,RIO with preference to short flows. Queue Management Policy – Drop Tail, RED,RIO with preference to short flows.

12. Link Utilization under Drop Tail, RED and RIO-PS

13. Network Goodput Under Different Schemes

14. Outline Introduction Introduction Analyzing Short TCP Flow Performance Analyzing Short TCP Flow Performance Architecture and Mechanism –RIO-PS Architecture and Mechanism –RIO-PS Simulations Simulations Discussions Discussions Conclusions and Future work Conclusions and Future work

15. Proposed Architecture Proposed Architecture

16. Edge Router Determines packet coming from long or short flow Determines packet coming from long or short flow Maintains a counter L t that tracks how many packets have been observed so far for a flow. L t is dynamic Maintains a counter L t that tracks how many packets have been observed so far for a flow. L t is dynamic Per flow state information are softly maintained to detect the termination of flow. The flow hash table is updated periodically every T u time units. Per flow state information are softly maintained to detect the termination of flow. The flow hash table is updated periodically every T u time units. It is configured with SLR (Short to Long ratio). It is configured with SLR (Short to Long ratio). It then periodically (every T c time units) performs AIAD control over the threshold to achieve the target SLR It then periodically (every T c time units) performs AIAD control over the threshold to achieve the target SLR

17. Core Router Gives preferential treatment to mice Gives preferential treatment to mice RIO (Red In and Out) queuing policy is used[4] with preferential treatment to short flows- RIO-PS RIO (Red In and Out) queuing policy is used[4] with preferential treatment to short flows- RIO-PS RIO used twin RED algorithms for dropping packets one for ins and one for outs. RIO used twin RED algorithms for dropping packets one for ins and one for outs. The probability of dropping “in” packets depends on the in average “in” packet queue and the probability of dropping “out” packets depend on the total average queue length. The probability of dropping “in” packets depends on the in average “in” packet queue and the probability of dropping “out” packets depend on the total average queue length. No packet reordering will happen in the FIFO queue with RIO No packet reordering will happen in the FIFO queue with RIO RIO inherits all features of RED RIO inherits all features of RED RIO performs soft prioritization, thus does not lose the benefit of statistical multiplexing. RIO performs soft prioritization, thus does not lose the benefit of statistical multiplexing.

18. RIO Queue with preferential treatment to short flows

19. Outline Introduction Introduction Analyzing Short TCP Flow Performance Analyzing Short TCP Flow Performance Architecture and Mechanism –RIO-PS Architecture and Mechanism –RIO-PS Simulations Simulations Discussions Discussions Conclusions and Future work Conclusions and Future work

20. Simulation Setup

21. Simulation Configuration

22. Experiment 1 4000 secs simulation time,2000 secs warm up time. 4000 secs simulation time,2000 secs warm up time. Average response time relative to RED Average response time relative to RED

23. Instantaneous Queue Size and Drop Rate

24. Fairness of Transmission time

25. Transmission Time of foreground traffic

26. Network goodput

27. Experiment 2:Unbalanced Requests Client set 1 requests smaller objects,Client set 2 requests larger objects

28. Experiment 2

29. Outline Introduction Introduction Analyzing Short TCP Flow Performance Analyzing Short TCP Flow Performance Architecture and Mechanism –RIO-PS Architecture and Mechanism –RIO-PS Simulations Simulations Discussions Discussions Conclusions and Future work Conclusions and Future work

30. Discussion Simulation Model Simulation Model Dumbbell and Dancehall model used. Dumbbell and Dancehall model used. All TCP connections have similar end to end propagation delays, this is not common topology seen by internet users All TCP connections have similar end to end propagation delays, this is not common topology seen by internet users If reverse traffic present even better performance If reverse traffic present even better performance Queue Management Policy Queue Management Policy RIO neither provides absolute aggregate (class based) nor relative flow based guarantees. RIO neither provides absolute aggregate (class based) nor relative flow based guarantees. Other AQM policies like PI controlled RED queue better Other AQM policies like PI controlled RED queue better

31. Discussions Deployment Issues Deployment Issues Edge devices need to perform per-flow state maintenance and per packet processing but it does not effect performance. Edge devices need to perform per-flow state maintenance and per packet processing but it does not effect performance. Not required to implement queue policies at each router, RIO-PS can be implemented at busy bottleneck links. Not required to implement queue policies at each router, RIO-PS can be implemented at busy bottleneck links. Flow Classification Flow Classification Threshold based classification classifies the first few packets of all flows to be short but it helps enhance performance. Threshold based classification classifies the first few packets of all flows to be short but it helps enhance performance.

32. Discussions Controller design Controller design The actual SLR depends on values of Tc and Tu, which determines Lt. Smaller values of these increases accuracy at the expense of increased overhead The actual SLR depends on values of Tc and Tu, which determines Lt. Smaller values of these increases accuracy at the expense of increased overhead Malicious users Malicious users Can break long transmissions into short flows but overhead of fragmentation and reassembly is very high. Can break long transmissions into short flows but overhead of fragmentation and reassembly is very high.

33. Outline Introduction Introduction Analyzing Short TCP Flow Performance Analyzing Short TCP Flow Performance Architecture and Mechanism –RIO-PS Architecture and Mechanism –RIO-PS Simulations Simulations Discussions Discussions Conclusions and Future work Conclusions and Future work

34. Conclusions and Future Work Performance of mice is improved Performance of mice is improved Performance of few elephants is also improved Performance of few elephants is also improved Overall goodput of the system is also improved Overall goodput of the system is also improved The proposed architecture is flexible in that the functionality that defines this scheme can be largely tuned at the edge routers The proposed architecture is flexible in that the functionality that defines this scheme can be largely tuned at the edge routers

35. Future work Integrate size aware traffic management at both network and transport layers Integrate size aware traffic management at both network and transport layers