Robust TCP congestion recovery

This paper presents a new robust TCP congestion-recovery scheme to (1) handle bursty packet losses while preserving the self-clocking capability; (2) detect a TCP connection's new equilibrium during congestion recovery, thus improving both link utilization and effective throughput; and (3) make the TCP behavior during congestion recovery very close to that during congestion avoidance, thus "extending" the performance model for congestion avoidance to that for TCP loss recovery. Moreover, the new recovery scheme requires only a slight modification to the sender side of TCP implementation, thus making it widely deployable. The performance of the proposed scheme is evaluated for scenarios with many TCP flows under the drop-tail and RED gateways in the presence of bursty packet losses. The evaluation results show that the new scheme achieves at least as much performance improvements as TCP SACK and consistently outperforms TCP New-Reno. Furthermore, its steady-state TCP behavior is close to the ideal TCP congestion behavior. Since the proposed scheme does not require selective acknowledgments nor receiver modifications, its implementation is simpler than SACK, and only the servers in the Internet need to be modified while keeping intact millions of clients scattered in the Internet.