Predicting TCP throughput from non-invasive network sampling

Proceedings - IEEE INFOCOM - Tập 1 - Trang 180-189 vol.1
M. Goyal1, R. Guerin2, R. Rajan3
1Computer &Information Science Department, Ohio State Uinversity, Columbus, OH, USA
2Department of Electrical Engineering, University of Pennsylvania, Philadelphia, PA, USA
3CTO, Ipsum Networks, Philadelphia, PA, USA

Tóm tắt

In this paper, we wish to derive analytic models that predict the performance of TCP flows between specified endpoints using routinely observed network characteristics such as loss and delay. The ultimate goal of our approach is to convert network observables into representative user and application relevant performance metrics. The main contributions of this paper are in studying which network performance data sources are most reflective of session characteristics, and then in thoroughly investigating a new TCP model based on Padhye et al. (2000) that uses non-invasive network samples to predict the throughput of representative TCP flows between given end-points.

Từ khóa

#Throughput #Sampling methods #Performance loss #Performance analysis #Monitoring #Application software #Predictive models #Scalability #Delay estimation #Computer networks

Tài liệu tham khảo

0, The network simulator - NS2 mathis, 1996, TCP selective acknowledgement options 10.1145/505672.505674 10.1109/90.251892 mccloghrie, 1991, Management information base for network management of TCP/IP-based internets: MIB-II jacobson, 0, Berkeley TCP evolution from 4.3-Tahoe to 4.3-Reno, Proceedings of the Eighteenth Internet Engineering Task Force Vancouver B C 1990, 365 goyal, 2001, Predicting TCP throughput from non-invasive data paxson, 2000, Computing TCP's retransmission timer, 10.17487/rfc2988 wright, 1996, TCP/IP Illustrated Volume 2 The Implementation 10.1109/90.611099 10.1145/251007.251012 allman, 1999, TCP congestion control, 10.17487/rfc2581 10.1109/90.842137 10.1109/90.909022 10.1145/301453.301550 altman, 0, A stochastic model of TCP/IP with stationary random losses, Proceedings of ACM SIGCOMM 2000 2000 10.1109/90.720921 10.1145/263932.264023 10.1109/INFCOM.2000.832574 10.1109/INFCOM.2000.832572