<DIV>I have not find any material related to how Netli's protocol</DIV>
<DIV>works to avoid the slow-start procedure in TCP connections, </DIV>
<DIV>how their congestion control is adopted, and how fairness is achived.</DIV>
<DIV> </DIV>
<DIV> The related page is:</DIV>
<DIV> </DIV>
<DIV><A href="http://www.nwfusion.com/cgi-bin/forum/gforum.cgi?do=post_view_threaded;post=370;sb=post_latest_reply;so=ASC;t=default">http://www.nwfusion.com/cgi-bin/forum/gforum.cgi?do=post_view_threaded;post=370;sb=post_latest_reply;so=ASC;t=default</A></DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV><BR><B><I>"Armando L. Caro Jr." <acaro@mail.eecis.udel.edu></I></B> wrote:</DIV>
<BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid">Not to re-start an old (and long) thread, but this mail is a side-note to<BR>one of Alex's mails...<BR><BR>On Mon Mar 31 22:57:08 2003, Alex Cannara wrote:<BR>> I found it interesting some years ago, when consulting for a large storage<BR>> apps company, that their TCP traffic was violating most TCP rules -- no<BR>> slow start, window shutdowns ignored... It was amazing to watch with a<BR>> Sniffer(r) the vast data rates they achieved among big Sun servers. I was<BR>> ready to list the violations as problems with their stacks when one of<BR>> their development guys explained that they knew enough about how the OS<BR>> handled IP that they didn't need to use the kludges added to TCP to<BR>> 'protect' the net. Many on this list have likely used this vendor's<BR>> systems in one way or another.<BR><BR>What worries me is how much praise such violations receive. Network
World<BR>(issue 04/21/03) has an article that praises a startup<BR>called Netli, which has developed a producted called NetLightning.<BR>NetLightning is a CDN architecture which essentially establishes three<BR>transport layer connections: (1) client to local VDC (Virtual Data<BR>Center), (2) local VDC to remote AAP (Application Access Point), and (3)<BR>remote AAP to HTTP server. Connections (1) and (3) use the normal<BR>HTTP/TCP, but connection (3) uses Netli's prioprietary transport protocol.<BR><BR>According to Netli, their transport protocol "uses the reliable, secure,<BR>and network-friendly techniques of TCP, but is enhanced to provide optimal<BR>performance for communication between a VDC in one geographic region and<BR>an AAP in another." However, Tim Greene of Network World reports more<BR>details which reveal just how network-friendly Netli's protocol is. For<BR>example, Greene reports "that the [Netli] protocol can reduce from 31 to<BR>two the number of round-trip
interactions needed to send a 70K byte Web<BR>page with 25 objects on it." Also, the Netli Protocol "eliminates the<BR>slow-start feature of TCP," because "these and other refinements of TCP<BR>reduce the time it takes to download pages and to interact with Web<BR>applications."<BR><BR>NetLightning could become a problem if it becomes deployed as widely as<BR>Alkami's CDN technology. With customers such as HP and investors such as<BR>Nokia, it is likely that NetLightning will spread and Netli's competitors<BR>will adopt similar techniques.<BR><BR>FYI: Netli are present at the Networld+Interop in Las Vegas, and they have<BR>a marketing talk today at 2pm PST.<BR><BR>Network World Article:<BR>http://www.nwfusion.com/news/2003/0421netli.html<BR><BR>Netli's Website:<BR>www.netli.com<BR><BR>~armando<BR><BR>0-- --0<BR>| Armando L. Caro Jr. | Protocol Engineering Lab |<BR>| www.cis.udel.edu/~acaro | University of Delaware |<BR>0-- --0<BR></BLOCKQUOTE><BR><BR>Jing Shen<br><br>State Key Lab
of CAD&CG<br>ZheJiang University(YuQuan)<br>HangZhou, ZheJiang Province 310027<br>P.R.China<p><br><hr size=1><b>Do You Yahoo!?</b><br>
<a href="http://cn.rd.yahoo.com/mail_cn/tag/?http://cn.surveys.yahoo.com/netlove">ÍøÁµµÄÚ¹ÊÍ£ºÕæÇ黹ÊÇ·Å×Ý£¿</a>