FAST TCP: motivation, architecture, algorithms, performance

Cheng Jin, David X. Wei, Steven H. Low

2005 · 628 citations · 49 references

Concepts

TL;DR

FAST TCP is a new congestion‑control algorithm designed to rapidly stabilize high‑speed, long‑latency networks into steady, efficient, and fair operating points in dynamic sharing environments. FAST TCP addresses four packet‑ and flow‑level difficulties of large windows, implements a novel architecture, and achieves equilibrium and stability properties, as demonstrated in a Linux prototype that was compared to TCP Reno, HSTCP, and STCP. Experimental results show that FAST TCP outperforms TCP Reno, HSTCP, and STCP in throughput, fairness, stability, and responsiveness, with promising preliminary performance.

Abstract

We describe FAST TCP, a new TCP congestion control algorithm for high-speed long-latency networks, from design to implementation. We highlight the approach taken by FAST TCP to address the four difficulties, at both packet and flow levels, which the current TCP implementation has at large windows. We describe the architecture and characterize the equilibrium and stability properties of FAST TCP. We present experimental results comparing our first Linux prototype with TCP Reno, HSTCP, and STCP in terms of throughput, fairness, stability, and responsiveness. FAST TCP aims to rapidly stabilize high-speed long-latency networks into steady, efficient and fair operating points, in dynamic sharing environments, and the preliminary results are promising.

References

49