Connections with multiple congested gateways in packet-switched networks part 1: one-way traffic
ACM SIGCOMM Computer Communication Review
Wide area traffic: the failure of Poisson modeling
IEEE/ACM Transactions on Networking (TON)
Modeling TCP Reno performance: a simple model and its empirical validation
IEEE/ACM Transactions on Networking (TON)
Fluid-based analysis of a network of AQM routers supporting TCP flows with an application to RED
Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
Statistical bandwidth sharing: a study of congestion at flow level
Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications
A mean-field model for multiple TCP connections through a buffer implementing RED
Performance Evaluation
Global stability of internet congestion controllers with heterogeneous delays
IEEE/ACM Transactions on Networking (TON)
Poster: TCP effective bandwidth and internet performance
ACM SIGMETRICS Performance Evaluation Review - Special Issue on IFIP PERFORMANCE 2011- 29th International Symposium on Computer Performance, Modeling, Measurement and Evaluation
How to model a TCP/IP network using only 20 parameters
Proceedings of the Winter Simulation Conference
Hi-index | 0.00 |
We propose an improved model for TCP flow level congestion performance. Conventional fluid approximation models yield incorrect results for TCP congestion at the flow level due to throughput overestimation. We introduce a more accurate model incorporating some of the packet level burstiness, by approximating TCP sources as on-off fluid sources. Our model is largely based on the work of D. Anick, D. Mitra and M.M. Sondhi exploring buffer statistics of the superposition of on-off sources. Incorporating some of the TCP burstiness at the packet level while retaining the over all fluid approximation framework allows us to substantially improve the accuracy of the flow level congestion model at the price of a relatively small increase in mathematical complexity. The model is extensively validated against ns2 simulations and shown to perform better than the M/M/1/B based model typically used in this context.