[37240] in cryptography@c2.net mail archive
Re: Hamiltonian path as protection against DOS.
daemon@ATHENA.MIT.EDU (James A. Donald)
Sun Aug 20 20:38:27 2006
X-Original-To: cryptography@metzdowd.com
X-Original-To: cryptography@metzdowd.com
Date: Fri, 18 Aug 2006 06:58:06 +1000
From: "James A. Donald" <jamesd@echeque.com>
To: Anne & Lynn Wheeler <lynn@garlic.com>
Cc: michaelslists@gmail.com, Bill Stewart <bill.stewart@pobox.com>,
cryptography@metzdowd.com
In-Reply-To: <44E375EA.2000200@garlic.com>
--
Anne & Lynn Wheeler wrote:
> as an aside, i've pointed out before that in the
> mid-90s that as webserver activity was increasing ...
> a lot of platforms experienced severe throughput
> degradation with HTTP transaction protocol use of TCP.
> Most platforms had a highly inefficient session close
> implementation around checking of the FINWAIT list ...
> the assumption as that most session activity had
> relatively infrequent session open/close activity. The
> HTTP transaction activity violated those TCP activity
> assumptions ... and for a period of time you found
> platforms spending over 95percent of their processor
> utilization dealing with the FINWAIT list.
Has this been entirely fixed, or is some substantial
degree of inefficiency inherent in the TCP protocol if
one uses it in a manner that tends to approximate UDP.
(Lots of brief connections that are predominantly one
way.)
The keep-alive extension to the HTTP protocol was
intended to make HTTP activity more TCP like.
--digsig
James A. Donald
6YeGpsZR+nOTh/cGwvITnSR3TdzclVpR0+pr3YYQdkG
8PdgpCY8lcy6qRnKRQ4rc2g1XHLHgfmlDh2ajbn/
4Tdc/z1dVOW8Pb51y7ZwS1xLayi1u3YmFU8aAvRdv
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo@metzdowd.com
