A TCP With Guaranteed Performance in Networks with Dynamic Congestion and Random Wireless Losses Stefan Schmid, ETH Zurich Roger Wattenhofer, ETH Zurich 2nd Annual International Wireless Internet Conference (WICON) Boston, MA, USA, August 2006 Distributed Computing Group

Stefan Schmid, ETH WICON Large Data Transfers (1) CERN, Geneva ETH, Zurich

Stefan Schmid, ETH WICON Large Data Transfers (2) CERN, Geneva ETH, Zurich TCP Connection Lecture E = mc 2

Stefan Schmid, ETH WICON Large Data Transfers (3) Characteristics of transfer: - Internet can be congested - Available bandwidth changes over time - Packets may be lost, especially on wireless link Congestion Losses

Stefan Schmid, ETH WICON TCP Congestion Control (1) TCP avoids congestion - Congestion control lies at the heart of TCP - Prevents congestion collapses of Internet (e.g., 1980) How to prevent congestion? Senders reduce sending rate when Internet is congested - senders maintain congestion window - strategy: „additive-increase, multiplicative-decrease“ (AIMD)

Stefan Schmid, ETH WICON TCP Congestion Control (2) How does a sender know about congestion? When packets are lost, TCP sender assumes that routers are overloaded! For packets lost for other reasons than congestion, throughput is reduced unnecessarily!

Stefan Schmid, ETH WICON TCP Congestion Control (3) Lecture E = mc 2 Wasted throughput: student may seek to increase her download bandwidth (selfishly)!

Stefan Schmid, ETH WICON In this paper… A model is presented which comprises both dynamic changes of the available bandwidth (dynamic congestion) and random packet losses (e.g., wireless links) Model allows for formal analysis of transfer protocol‘s performance! Thereby, a selfish perspective is assumed - We look at protocols which aim at maximizing their throughput, regardless of consequences for other participants (no fairness).

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Basic Model (1) Time is divided into synchronous rounds Framework of online algorithms: - Adversary chooses available bandwidth u t - Protocol chooses sending rate x t In addition, all packets are lost in a given round with probability p

Stefan Schmid, ETH WICON Basic Model (2) Gain of transfer protocol ALG at time t: Gain of optimal (offline) transfer strategy OPT: We are interested in minimizing the (strict) competitive ratio, i.e., the gain of OPT divided by the gain of ALG.

Stefan Schmid, ETH WICON Basic Model (3) Takes into account an opportunity cost Assumption: No packets are transmitted at all if rate too large - Pessimistic, but losses engender overhead (e.g., time-outs) Goal of ALG is to always send at (or slightly lower) rate of currently available bandwidth. Thereby, ALG does not know whether losses are due to congestion or wireless links!

Stefan Schmid, ETH WICON In practice, it can be assumed that congestion does not change too abruptly over time. Therefore, we bound the adversary ADV which chooses the available bandwidth as follows (multiplicative changes): u t has to be chosen from [u t-1 /μ, u t-1 μ ] Basic Model (4) A similar model but without random losses has been studied by Karp, Koutsoupias, Papadimitriou and Shenker (FOCS 2000)!

Stefan Schmid, ETH WICON Transfer protocol achieving a provable performance: Provable Performance in Basic Model In order to compensate wireless losses, TCPW increases the bandwidth by a factor larger than μ after successful rounds (aggressive MIMD strategy). Strict competitive ratio: at most 4(μ 2 + μ)

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Network traffic is often bursty. Network calculus has introduced the notion of leaky-bucket arrival curves to study queuing theory from a worst-case perspective. Also a reasonable model for dynamics on transport layer! Extending the Model with Bursts leaky bucket arrival curve

Stefan Schmid, ETH WICON Dynamics of ADV has to correspond to leaky-bucket constraints New Dynamic Adversary where Adversary can accumulate power in some rounds to change available bandwidth more abruptly later!

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON The TCP Wichita Transfer Protocol for Bursty Environment After successful transmissions, rate is increased by a factor of 2 μ 2. x t >u t : Transmission rate compared to OPT can be described by Markov chain. - Only in case of random errors, TCPW loses ground / reduces too much! Analysis: Look at cases x>u (fail) and x<u (pot. success) individually.

Stefan Schmid, ETH WICON Analysis In rounds where x t >u t, TCPW has gain = 0! - But: TCPW does not miss much gain! - TCPW reduces its rate gemoetrically - TCPW never overshoots much TCP Wichita is 4(μ+Bμ 2 )-competitive against a bursty adversary if μ<(1-p)/4Bp. The following result can be shown:

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Simulation Random bandwidth changes with bursts - Random changes smaller than μ, until enough is accumulated for burst

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Talk Overview Basic Model Extending the Model to Incorporate Bursts TCP Wichita and Analysis Simulation Conclusion

Stefan Schmid, ETH WICON Conclusion (1) A framework which allows for formal protocol analysis and incorporates dynamic congestion and random losses We believe that there is still little algorithmic research on the transport layer! Network calculus may be a good model for dynamics in various settings! Selfish throughput maximization - Really a threat? Experiences? - Security: Routers often drop UDP packets first in case of congestion! Cheating possible?

Stefan Schmid, ETH WICON Conclusion (2) Open research questions: - Better / tight bound for competitive ratio? - Randomized online algorithms? - Impact on stability in case of multiple flows? - Model extensions: buffers, varying round trip times, etc.? - Model verification & real implementation?

Stefan Schmid, ETH WICON Questions and Comments? Stefan Schmid Distributed Computing Group ETH Zurich, Switzerland Thank you for your attention!