Date: Sun, 01 Feb 2004 14:14:03 -0600 From: Jeremy Messenger <mezz7@cox.net> To: Mike Silbersack <silby@silby.com> Cc: cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/kern kern_subr.c uipc_mbuf.cuipc_syscalls.c src/sys/sys uio.h Message-ID: <opr2pjtpkk8ckrg5@smtp.central.cox.net> In-Reply-To: <20040201140055.C3097@odysseus.silby.com> References: <200402010756.i117uiWm094818@repoman.freebsd.org> <20040201085037.GA15540@regency.nsu.ru> <20040201140055.C3097@odysseus.silby.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 1 Feb 2004 14:02:11 -0600 (CST), Mike Silbersack <silby@silby.com> wrote: > > On Sun, 1 Feb 2004, Alexey Dokuchaev wrote: > >> > Rewrite sendfile's header support so that headers are now sent in >> the first >> > packet along with data, instead of in their own packet. When >> serving files >> > of size (packetsize - headersize) or smaller, this will result in >> one less >> > packet crossing the network. Quick testing with thttpd and >> http_load has >> > shown a noticeable performance improvement in this case (350 vs 330 >> fetches >> > per second.) >> >> Good news to us thttpd users out there. :-) >> Thanks! >> >> ./danfe > > Heh, that reminds me... With the same test parameters, Apache2 gets a > whopping 220 fetches per second. Go thttpd! :) > > (Hopefully once we enable threading support in apache2 it will go > faster... hopefully.) Have one of you tried AppWeb[1] yet? I learned about it from the Slashdot[2] and it claimed that it outperforms Apache that can deliver about 3.5K pages per second on a Pentium-class system. [1] http://www.mbedthis.com [2] http://apache.slashdot.org/article.pl?sid=04/01/27/2210229&mode=thread&tid=148&tid=185 Cheers, Mezz > Mike "Silby" Silbersack -- bsdforums.org 's moderator, mezz.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?opr2pjtpkk8ckrg5>