Messages in this thread | | | From | Josh Cohen <> | Subject | RE: my broken TCP is faster on broken networks | Date | Sat, 12 Sep 1998 13:52:35 -0700 |
| |
> -----Original Message----- > From: alan@lxorguk.ukuu.org.uk [mailto:alan@lxorguk.ukuu.org.uk] > Sent: Saturday, September 12, 1998 7:01 AM > To: Josh Cohen > Cc: marcs@znep.com; linux-kernel@vger.rutgers.edu > Subject: Re: my broken TCP is faster on broken networks > > > Charles Hannum pretty much buried the original T/TCP from its security > viewpoint. The rather interesting beginnings of the "how do we rewrite > T/TCP" discussion disappeared off tcp-impl in the direction of end2end > very recently too. > agreed.
Along similar lines, and a point I meant to emphasize but forgot to, is looking at the handshake. Looking at the user abort cases, the loss of the first SYN is extremely painful. Network traces show that on many stacks, SYN retransmission can occure at 6 seconds and 24 seconds. Essentially, if the first SYN of the handshake is lost, you may wait 6 seconds before it is even retransmitted, often this is enough to make the user hit stop/reload. One of the nice points of T/TCP was that it tried to reduce the handshake time. Even though T/TCP is all but dead, it is interesting to consider the SYN with data proposals. These days it seems like we spend more time in the connection startup and teardown than in data transfer :)
Josh
/* Disclaimer: The opinions expressed here are my own and not those of my employer. This post, in particular, represents nothing more than a personal interest in this discussion and does not imply any actions, planned actions, or intent on the part of my employer. */
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/faq.html
| |