<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:times new roman, new york, times, serif;font-size:10pt"><DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif">Lachlan,</DIV>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif"> </DIV>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif">I've parsed your reference. I guess your focus is on "application constrained/idle" scenario.</DIV>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif">Of course, in this case, there is no ambiguity about RTO causes...</DIV>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif"> </DIV>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif">Rgds,</DIV>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif">Dirceu<BR><BR></DIV>
<DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif">----- Original Message ----<BR>From: Dirceu Cavendish <dirceu_cavendish@yahoo.com><BR>To: Lachlan Andrew <lachlan.andrew@gmail.com><BR>Cc: "iccrg@cs.ucl.ac.uk" <iccrg@cs.ucl.ac.uk><BR>Sent: Friday, March 28, 2008 12:09:37 PM<BR>Subject: Re: [Iccrg] SSDT Scope - summary<BR><BR>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif">
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif">Lachlan,<BR><BR>I will look at your reference. But taking the risk of a quick reply before that...<BR><BR>One problem I see is how we can differentiate between RTOs for which the path does not change from the ones that are caused by path change...<BR><BR>My problem with Quick Start is its required support from routers, before anything else.<BR><BR>Any solution will have to strike a balance between being too aggressive in certain network scenarios and not being aggressive enough in others. Unless the solution is adaptive...<BR><BR>But definitely, multiple SS issue should be listed as a separate one, although it may be triggered by overshooting of the buffer at large BDP, already listed...<BR><BR>Thanks for pointing that out.<BR>Dirceu<BR><BR><BR>
<DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif">----- Original Message ----<BR>From: Lachlan Andrew <lachlan.andrew@gmail.com><BR>To: Dirceu Cavendish <dirceu_cavendish@yahoo.com><BR>Cc: "iccrg@cs.ucl.ac.uk" <iccrg@cs.ucl.ac.uk><BR>Sent: Friday, March 28, 2008 10:51:09 AM<BR>Subject: Re: [Iccrg] SSDT Scope - summary<BR><BR>Greetings Dirceu,<BR><BR>One "low-hanging fruit" would be to fix a slow start which occurs<BR>after timeout or an idle period, rather than at the start of a<BR>connection. In these cases, the sender has a good guess at the right<BR>BDP. In Chicago, Sally gave a presentation suggestion quadrupling<BR>(instead of doubling) up to the guessed BDP. That helps problem 1,<BR>but exacerbates problem 2.<BR><BR>If we have an estimate of the BDP, then *both* problems can be solved<BR>by increasing the rate linearly (at a suitable rate) rather than<BR>exponentially. Take
a look at<BR><<A href="http://netlab.caltech.edu/lachlan/abstract/AIslowstart.pdf" target=_blank rel=nofollow>http://netlab.caltech.edu/lachlan/abstract/AIslowstart.pdf</A>> if you're<BR>interested.<BR><BR>Personally, I'm not a fan of QuickStart/JumpStart, because it doesn't<BR>give other flows enough time to back off. However, using QuickStart<BR>signalling to specify the final BDP would allow this linear increase<BR>over a suitable number of RTTs for slow start at the start of a<BR>connection too.<BR><BR>$0.02,<BR>Lachlan<BR><BR>On 28/03/2008, Dirceu Cavendish <<A href="mailto:dirceu_cavendish@yahoo.com" target=_blank rel=nofollow ymailto="mailto:dirceu_cavendish@yahoo.com">dirceu_cavendish@yahoo.com</A>> wrote:<BR>><BR>> I went through this week's messages on SS topics, and here is a summary of<BR>> them:<BR>><BR>> - Transaction latencies taking too long to complete, in a large bandwidth<BR>> delay product
networks.<BR>> - Buffer overshoot, causing multiple segment losses/retransmits, potentially<BR>> causing RTOs<BR>><BR>> The first item can be seen as SS being too conservative/slow, whereas the<BR>> second one may be cast as SS being too aggressive in doubling cwnd when<BR>> current cwnd is large (high BDP networks). So, my proposal would be to scope<BR>> the SSDT as to CHARACTERIZE these two issues first. Depending on how fast we<BR>> go with this work, we can then address/analyze solutions, such as Quick<BR>> Start, limited slow start, and others...But I think a good characterization<BR>> will bring already the features we would like to see in a solution, without<BR>> having to lock ourselves up to any one.<BR>><BR>> Dirceu<BR>><BR>><BR>><BR>> ________________________________<BR>> Looking for last minute shopping deals? Find them fast with Yahoo! Search.<BR>>
_______________________________________________<BR>> Iccrg mailing list<BR>> <A href="mailto:Iccrg@cs.ucl.ac.uk" target=_blank rel=nofollow ymailto="mailto:Iccrg@cs.ucl.ac.uk">Iccrg@cs.ucl.ac.uk</A><BR>> <A href="http://oakham.cs.ucl.ac.uk/mailman/listinfo/iccrg" target=_blank rel=nofollow>http://oakham.cs.ucl.ac.uk/mailman/listinfo/iccrg</A><BR>><BR>><BR><BR><BR>-- <BR>Lachlan Andrew Dept of Computer Science, Caltech<BR>1200 E California Blvd, Mail Code 256-80, Pasadena CA 91125, USA<BR>Ph: +1 (626) 395-8820 Fax: +1 (626) 568-3603<BR><A href="http://netlab.caltech.edu/lachlan" target=_blank rel=nofollow>http://netlab.caltech.edu/lachlan</A><BR></DIV><BR></DIV></DIV><BR>
<HR SIZE=1>
Looking for last minute shopping deals? <A href="http://us.rd.yahoo.com/evt=51734/*http://tools.search.yahoo.com/newsearch/category.php?category=shopping" target=_blank rel=nofollow>Find them fast with Yahoo! Search.</A></DIV>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif"><BR></DIV></div><br>
<hr size=1>Special deal for Yahoo! users & friends - <a href="http://us.rd.yahoo.com/evt=47521/*http://tc.deals.yahoo.com/tc/blockbuster/text3.com
">No Cost. Get a month of Blockbuster Total Access</a> now</body></html>