<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=ks_c_5601-1987">
<META content="MSHTML 5.50.4611.1300" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2>Hi,</FONT></DIV>
<DIV><FONT face=Arial size=2> For the 1st question, is the
assumption sound?</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2> For the second question, the major
difficulty is FQ is time consuming. In order to do FQ, you should do
</FONT></DIV>
<DIV><FONT face=Arial size=2> 1 flow identification. For IP packet,
possibly you need to take into consideration the IP dest/src address, TCP port
number, TOS, etc. </FONT></DIV>
<DIV><FONT face=Arial size=2> 2 scheduling, which requires that the
router must do some sort to determine the output sequence.</FONT></DIV>
<DIV><FONT face=Arial size=2> These 2 steps take some time. You can
check the website of Juniper Networks Inc. and see how they are trying solve
this problem. </FONT></DIV>
<DIV><FONT face=Arial size=2> Regards.</FONT></DIV>
<DIV><FONT face=Arial size=2> </FONT></DIV>
<DIV><FONT face=Arial size=2> Yan Wu</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV>----------------------------------------------------------------------</DIV>
<DIV><FONT face=Arial size=2> Telecommunication Research
Center </FONT></DIV>
<DIV><FONT face=Arial size=2> Department of Electrical
Engineering
</FONT></DIV>
<DIV><FONT face=Arial size=2> Arizona State
University </FONT></DIV>
<DIV><FONT face=Arial size=2> </FONT></DIV>
<BLOCKQUOTE dir=ltr
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV
style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B>
<A title=ggumdol@comis.kaist.ac.kr
href="mailto:ggumdol@comis.kaist.ac.kr">Jeong-woo Cho</A> </DIV>
<DIV style="FONT: 10pt arial"><B>To:</B> <A title=end2end-interest@postel.org
href="mailto:end2end-interest@postel.org">End2End</A> </DIV>
<DIV style="FONT: 10pt arial"><B>Sent:</B> Wednesday, August 22, 2001 9:00
AM</DIV>
<DIV style="FONT: 10pt arial"><B>Subject:</B> [e2e] Fundamental Questions
about Router Queue in High Speed IP Networks</DIV>
<DIV><FONT face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT
face=Arial size=2></FONT><BR></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2> Studying scheduling algorithms(FQ, SFQ, RED,
FRED, REM, BRED, SRED) in IP networks, I got some questions that I cannot
answer. If you have any ideas on the following questions, please let me know
the answers or the pointers to related literature.</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>** Assumptions : We can reduce larger queuing delays caused
by huge router queue size by using several schemes.</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>1. Why do routers have small queue sizes?</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2> Because TCP is a window-based congestion control
scheme, If routers provide small queue sizes such as 60 kbytes or 90 kbtytes,
only up to 40 or 60 TCP connections can be buffered in a router queue. I think
that we should increase router queue sizes to support more TCP connections
without TCP's coarse retransmit timeout and to support high value of fairness
because TCP's VERY coarse timeout itself induces unfairness. Is there any
technological obstables for providing larger router queue sizes? To support
thoudands of TCP connections in core routers, I think that router
should provide TCP connections with several giga bytes queue
sizes.</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>2. Why do researchers stress on Single FIFO scheduling
schemes such as RED?</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2> If core routers can provide several giga bytes queue
size, why don't core routers use Fair Queuing schemes such as Fair Queuing and
Stochastic Fair Queuing? I think RED itself and its variant are far
to support minimum QoS such as fairness. RED is unscalable in that RED can't
prevent buffer overflows when there are many TCP connections. Although
SRED(Stabilized RED) solved this problem, I do not think that it can
protect TCP connections from unresponsive flows and can provide fairness even
when there are only TCP connections.</FONT></DIV>
<DIV><FONT size=2> I think that FQ should be implemented in routers, if
per-flow queuing can be implemented. Is there any technological obstables for
implementing per-flow queuing in high speed core routers?</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2>Thanks in advance.</FONT></DIV>
<DIV><FONT size=2></FONT> </DIV>
<DIV><FONT size=2> </DIV>
<DIV><BR>----------------------------------------------------------------------<BR> Cho,
Jeong-woo<BR> <BR> Communication and Information Systems
Laboratory<BR> Dept. of Electrical Engineering and Computer
Science<BR> Korea Advanced Institute of Science and
Technology(KAIST)<BR> 373-1 Kusong-dong, Yusong-gu, Taejon 305-701,
Korea<BR> <BR> TEL: +82-42-869-8067 (ex.107) FAX:
+82-42-867-0550<BR> E-mail: <A
href="mailto:ggumdol@comis.kaist.ac.kr">ggumdol@comis.kaist.ac.kr</A><BR>----------------------------------------------------------------------<BR></DIV></BLOCKQUOTE></FONT></BODY></HTML>