<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"><P>Eddy,</P>
<P> </P>
<P>I will not dwell on VJ "intention" regarding fairness issues. However, from a control theoretical perspecive, it is a fact that a ON-OFF type of rate regulation embedded in the AIMD feature of TCP CA causes bottleneck bandwidth resharing, converging* to similar transmission rates.</P>
<P> </P>
<P>*I should be careful about the term convergence, as theoreticcally an ON-OFF controller never converges, but in steady state oscillates around a "fair" point. The oscillation band can be related with teh session rtt...</P>
<P> </P>
<P>Dirceu</P>
<DIV style="FONT-SIZE: 10pt; FONT-FAMILY: times new roman, new york, times, serif"><BR><BR>
<DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif">----- Original Message ----<BR>From: "Eddy, Wesley M. (GRC-RCN0)[VZ]" <wesley.m.eddy@nasa.gov><BR>To: Michael Welzl <michael.welzl@uibk.ac.at>; iccrg <iccrg@cs.ucl.ac.uk><BR>Sent: Thursday, April 17, 2008 6:11:10 AM<BR>Subject: RE: [Iccrg] Heresy recapped<BR><BR>>-----Original Message-----<BR>>From: <A href="mailto:iccrg-bounces@cs.ucl.ac.uk" ymailto="mailto:iccrg-bounces@cs.ucl.ac.uk">iccrg-bounces@cs.ucl.ac.uk</A> <BR>>[mailto:<A href="mailto:iccrg-bounces@cs.ucl.ac.uk" ymailto="mailto:iccrg-bounces@cs.ucl.ac.uk">iccrg-bounces@cs.ucl.ac.uk</A>] On Behalf Of Michael Welzl<BR>>Sent: Thursday, April 17, 2008 1:38 AM<BR>><BR>>>I will just point out that fairness is not a goal of VJ CC; avoiding<BR>>>packet loss is.<BR>>><BR>>>The inputs to VJ CC are "some bytes were not lost" (an ACK) and "some<BR>>>bytes were
probably lost" (an RTO or 3-dupack). They are not "someone<BR>>>is sending faster than me" or "I'm sending faster than someone else",<BR>>>or any other metric or inference useful for acheiving any form of<BR>>>fairness, because that is not its goal.<BR>><BR>>Now that ain't true. The choice of AIMD in the Cong. Avoid & Control<BR>>paper is based on the Chiu/Jain paper (this paper is cited in Van's),<BR>>where AIMD was found to be the only control leading to efficiency<BR>>*and fairness* among the four simple linear ones that they looked at.<BR><BR><BR>The message at:<BR><A href="http://ee.lbl.gov/tcp.html#dynamic" target=_blank>http://ee.lbl.gov/tcp.html#dynamic</A><BR>only briefly mentions fairness, and actually labels it as something<BR>for the router to decide, not the end hosts.<BR><BR>The from:<BR><A href="http://ee.lbl.gov/papers/congavoid.pdf"
target=_blank>http://ee.lbl.gov/papers/congavoid.pdf</A><BR>"While algorithms at the transport endpoints can insure the network<BR>capacity isn't exceeded, they cannot insure fair sharing of that<BR>capacity. Only in gateways, at the convergence of flows, is there enough<BR>information to control sharing and fair allocation."<BR><BR>Both references describe the selection of AIMD without reference to<BR>fairness, but specifically for stability.<BR><BR>_______________________________________________<BR>Iccrg mailing list<BR><A href="mailto:Iccrg@cs.ucl.ac.uk" 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>http://oakham.cs.ucl.ac.uk/mailman/listinfo/iccrg</A><BR></DIV><BR></DIV></div><br>
<hr size=1>Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. <a href="http://us.rd.yahoo.com/evt=51733/*http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ "> Try it now.</a></body></html>