[Iccrg] What's wrong with TCP Congestion control

S. Keshav keshav at uwaterloo.ca
Tue Feb 7 00:51:12 GMT 2006


Here is a list paraphrased from mail so far:

1. Unsustainable large equilibrium window sizes under high bandwidth-delay
product environments; requires an unrealistically low loss probability.
(Leads to limited dynamic range)
 2. Low throughput under lossy environments because of using loss as an
indication of congestion.
 3. Slow additive increase: Takes a long time for flow to catch up with
spare capacity and results in unnecessary long flow-completion times.
 4. Inefficient Slow-start: Even when the flow is capable of completing
within a round-trip time, slow-start makes flows last multiple round-trip
times just to find their fair share rate. Often most flows complete
before they exit slow-start phase.  
 5. Large queueing delay: TCP fills up any amount of buffering available at
the bottleneck links. Results in long latency. 
 6. Unfair bandwidth sharing: shares bandwidth inversely proportional to
flow RTTs
7. TCP builds a standing queue at the point of congestion, which increases
the delay. 

Did I miss anything?

thanks
keshav





More information about the Iccrg mailing list