Michael, In PR-SCTP where retranmission is set off, does it allows the applications to know which part of data is lost in the stream? Thanks
-gavin On 12/7/06, Michael Tuexen <[EMAIL PROTECTED]> wrote:
Hi Andre, see my comments in-line. Best regards Michael On Dec 7, 2006, at 10:01 AM, Andre Oppermann wrote: > [EMAIL PROTECTED] wrote: >> At Wed, 6 Dec 2006 23:09:39 +0800, >> maillist ifiaas wrote: >>> Hi friends, >>> >>> This is one of my research project. Our purpose is to modify TCP to >>> support unreliable but congestion controlled streaming. The >>> motivation is pretty similar to the one of DCCP CCID2. We have >>> implemented a prototype on FreeBSD 5.4, and the the modifications >>> are limited mostly in tcp_input.c and tcp_output.c. Source code, a >>> paper about the design (under submission), and an Iperf modification >>> to test out TCP Urel, is provided on the following page: >>> www.comp.nus.edu.sg/~malin/ >>> >>> Our current stage is changing Urel into a single directional >>> streaming protocol, so taht it could be abosolutely fair with >>> default TCP Sack, in FreeBSD. But we found after all the >>> modifications (on single directional streaming), Urel generates less >>> ACK than normal Sack, making it under utilized when competing to TCP >>> Sack. About only 3 out of 10 tries, Urel take the same throughput as >>> Sack. The reason seems to lying in the Delay ACK code, in >>> tcp_input.c. Because, when we turn off the Delay ACK option, using >>> sysctl command, Urel and Sack play fairly. However after days of >>> looking at the code, we failed to find the secret... Therefore, I >>> turn to you, the specialists of the TCP code in FreeBSD. Hope you >>> can help us to find the bug of our code. Any suggesion, comments, is >>> appreciated. >>> >>> For details of how the code is implemented, how our experiment is >>> conducted, you may need to spend one or two hours to browse through >>> our paper, and the source code. >> How is this different from the recently integrated SCTP? > > It doesn't try to retransmit at all. A lost segment is lost and > resending it would be pointless for realtime content. On the other > hand you don't want to blast the network at a fixed rate and so > this protocol wants to use a congestion control algorithm to back > off when bandwidth gets scarce. I haven't looked at the details > yet but my initial guess would be that the actual TCP code isn't > the best starting point. TCP is too obsessed with retransmitting > if something got lost. SCTP has a extension called PR-SCTP, which is implemented in BSD and can be used to limit the number of retransmissions of a DATA chunk to 0. The service you mention above is therefore available in SCTP. > > -- > Andre > > _______________________________________________ > freebsd-net@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-net > To unsubscribe, send any mail to "[EMAIL PROTECTED]" >
_______________________________________________ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
