230 likes | 428 Views
Scalable TCP: Improving Performance in Highspeed Wide Area Networks. Paper by Tom Kelly Presented by Christoph Jechlitschek. The Problem. TCP performs badly with very large congestion windows Converges slowly Underutilizes available link capacity
E N D
Scalable TCP: Improving Performance in Highspeed Wide Area Networks Paper by Tom Kelly Presented by Christoph Jechlitschek
The Problem • TCP performs badly with very large congestion windows • Converges slowly • Underutilizes available link capacity • Example: A 1 Gbps flow needs 28 minutes to recover from a single packet loss
TCP congestion window update • cwnd cwnd + 1/cwnd; - if no loss was detected • cwnd cwnd/2; - if a loss was detected
Scalable TCP congestionwindow update • cwnd cwnd + 0.01; - if no loss was detected • cwnd cwnd – 0.125 * cwnd; - if a loss was detected
Analysis • Generalized form of update function: cwnd cwnd + a; if no loss detected cwnd cwnd – b * cwnd; if loss detected • What makes a = 0.01 and b = 0.125 a good choice?
Legacy connections • Traditional connections: - are not designed to use large windows - limited amount of buffers • Legacy window size lwnd = 16 • Legacy loss rate pl = 5.86 * 10-3
Response curve • Relates a and b • Choosing one fixes the other
Instantaneous rate variation • Avoid large instantaneous rate variations • Coefficient of variance depends on b only • Therefore b should be as small as possible
Convergence • Converge to a new bound as fast as possible • Large a and b speed up convergence • Contradicts with earlier statement
Decision • Balance variation with converge time • Choose a = 0.01 and b = 0.125
Gigabit kernel modifications • Increases size of kernel interface queues • Avoids unnecessary packet copying • Removes debug counters
Performance test 1 • 4 sender/receiver pairs transfer 2 GB files • Count number of transfers in 1200 seconds
Performance test 2 • Simulate Web traffic using standard TCP • Background bulk transfer with Scalable TCP
Conclusion • Simple sender-side modifications greatly improved performance • No modification to receiver needed • Negligible effect on standard TCP connections • Incrementally deployable