Congestion avoidance and control
SIGCOMM '88 Symposium proceedings on Communications architectures and protocols
Performance measures and scheduling policies in ring networks
IEEE/ACM Transactions on Networking (TON)
The performance of TCP/IP for networks with high bandwidth-delay products and random loss
IEEE/ACM Transactions on Networking (TON)
Comparative performance analysis of versions of TCP in a local network with a lossy link
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
Feedback Control of Dynamic Systems
Feedback Control of Dynamic Systems
Design, analysis, and implementation of DVSR: a fair high-performance protocol for packet rings
IEEE/ACM Transactions on Networking (TON)
IEEE Network: The Magazine of Global Internetworking
Hi-index | 0.00 |
Resilient Packet Ring (RPR), a novel IEEE 802-series MAC scheme over optical network for Metropontan Area Networks (MAN), provides flexible data efficient delivery and optical transmission reliability. However, RPR fairness scheme causes TCP performance degradation. In this paper, an analytical traffic model is developed to study the impact of congestion control of TCP under RPR AM fairness scheme. Basically, two possible ways may be able to improve this impact, to equip a buffer to accommodate the burst or to employ a robust active queue management (AQM). However, buffer requirement in RPR networks may be different for various network scenarios and AQM scheme should be better for RPR networks. Furthermore, our study shows that TCP works not well if queue management parameters are fixed at a value. Hence, a novel adaptive queue management scheme with RPR feedback information is proposed. Available bandwidth for each ingress-aggregated flow should be passed to its upstream nodes in RPR MAC scheme. Accordingly, queue management design parameters could be adaptively tuned to achieve better queueing behavior. Through detailed simulations in various scenarios, our proposed schemes are able to improve TCP performance obviously.