<HTML><BODY style="word-wrap: break-word; -khtml-nbsp-mode: space; -khtml-line-break: after-white-space; ">Here's my take on the questions put forth by Keshav. For now I will limit my answers to elastic traffic.<DIV><DIV><BLOCKQUOTE type="cite"></BLOCKQUOTE><BR><BLOCKQUOTE type="cite"><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "> A. What is the commonly agreed definition for congestion. (I will post</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">my suggestion later today). If we do not agree on a definition, then indices</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">are meaningless.</DIV></BLOCKQUOTE></DIV></DIV><DIV><BR class="khtml-block-placeholder"></DIV><DIV>* A network-centric definition of congestion: Congestion is anything that deviates significantly from an efficient and fair usage of the network resources --- mostly network bandwidth and buffers.</DIV><DIV>* User-centric definition: From an user's view point what really matters is how quickly does my flow finish, so congestion is long flow-completion times. Often, it isn't the per-packet latency that users care about, but just how fast the entire flow completes.</DIV><DIV><BR class="khtml-block-placeholder"><BLOCKQUOTE type="cite"><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><SPAN class="Apple-converted-space"> </SPAN>B. What's wrong with TCP's congestion control scheme. Does someone have</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">a concise summary that they can post here? I am sure this will be a</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">cut-and-paste job for someone who has written a paper on congestion control</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">recently :-)</DIV></BLOCKQUOTE><DIV><BR class="khtml-block-placeholder"></DIV><DIV>Here's one list:</DIV><DIV><BR class="khtml-block-placeholder"></DIV><DIV> 1. Unsustainable large equilibrium window sizes under high bandwidth-delay product environments; requires an unrealistically low loss probability.</DIV><DIV> 2. Low throughput under lossy environments because of using loss as an indication of congestion.</DIV><DIV> 3. Slow additive increase: Takes a long time for flow to catch up with spare capacity and results in unnecessary long flow-completion times.</DIV><DIV> 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. </DIV><DIV> 5. Large queueing delay: TCP fills up any amount of buffering available at the bottleneck links. Results in long latency. </DIV><DIV> 6. Unfair bandwidth sharing: shares bandwidth inversely proportional to flow RTTs</DIV><DIV><BR class="khtml-block-placeholder"></DIV><DIV> -Nandita</DIV><BR><BLOCKQUOTE type="cite"><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><BR></DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">Lets see if we can reach consensus on these two issues in a week's time.</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">These will also form parts 1 and 2 of the consensus document. I ask for your</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">co-operation in keeping the discussion to these two issues.</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><BR></DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">thanks</DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><BR></DIV><DIV style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">kesahv<SPAN class="Apple-converted-space"> </SPAN></DIV></BLOCKQUOTE></DIV><BR><BR><DIV> <P style="margin: 0.0px 0.0px 0.0px 0.0px"><FONT face="Helvetica" size="3" style="font: 12.0px Helvetica">Nandita Dukkipati</FONT></P> <P style="margin: 0.0px 0.0px 0.0px 0.0px">Graduate Student </P><P style="margin: 0.0px 0.0px 0.0px 0.0px">Computer Systems Lab</P> <P style="margin: 0.0px 0.0px 0.0px 0.0px"><FONT face="Helvetica" size="3" style="font: 12.0px Helvetica">Stanford University</FONT></P> <P style="margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica; min-height: 14.0px"><BR></P> <P style="margin: 0.0px 0.0px 0.0px 0.0px"><FONT face="Helvetica" size="3" style="font: 12.0px Helvetica">webpage: <A href="http://yuba.stanford.edu/~nanditad/">http://yuba.stanford.edu/~nanditad/</A></FONT></P> <P style="margin: 0.0px 0.0px 0.0px 0.0px; min-height: 14.0px"><FONT face="Helvetica" size="3" style="font: 12.0px Helvetica"><SPAN class="Apple-converted-space"> </SPAN></FONT></P> </DIV><BR></BODY></HTML>