From owner-freebsd-current Sat Aug 21 18:12:26 1999 Delivered-To: freebsd-current@freebsd.org Received: from mw4.texas.net (mw4.texas.net [206.127.30.14]) by hub.freebsd.org (Postfix) with ESMTP id EF183154DC for ; Sat, 21 Aug 1999 18:12:18 -0700 (PDT) (envelope-from rsnow@lgc.com) Received: from basil.dympna.COM (mnet07-19.sat.texas.net [209.99.53.19]) by mw4.texas.net (2.4/2.4) with ESMTP id UAA25188; Sat, 21 Aug 1999 20:12:02 -0500 (CDT) Received: from lgc.com (IDENT:rsnow@turbo [134.132.228.6]) by basil.dympna.COM (8.9.3/8.8.7) with ESMTP id UAA14148; Sat, 21 Aug 1999 20:12:02 -0500 (CDT) (envelope-from rsnow@lgc.com) Message-ID: <37BF4E63.9019DD62@lgc.com> Date: Sat, 21 Aug 1999 20:12:03 -0500 From: Rob Snow X-Mailer: Mozilla 4.61 [en] (X11; I; Linux 2.2.11 i686) X-Accept-Language: en MIME-Version: 1.0 To: Doug Cc: Matthew Dillon , "John S. Dyson" , current@FreeBSD.ORG Subject: Re: NFS HEADS UP (was Re: cvs commit: src/sys/nfs nfsm_subs.h xdr_subs.h) References: <199908211156.GAA20231@dyson.iquest.net.> <199908211649.JAA73759@apollo.backplane.com> <37BF4420.452C5076@gorean.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG I'm with you 100% I've got about 100 Bonnies of trying to optimize Linux Clients (pre and post NFSv3 patch) vs. FreeBSD server. I've gone to current so that I can watch it happen. I'm currently writing around 4.5M - 5.0M/sec vs. 10-15M/sec on my vinum stripe locally. I went through a similar thing 1.5years ago with IBM/Solaris/Irix servers that were sharing VOB's, Irix to Solaris was DOG slow and we finally conquered it wit a bunch of fiddling and patching. -Rob Doug wrote: > > Matthew Dillon wrote: > > > Does anyone know why our NFS clients are sending a separate RPC for each > > 8K buffer? If the dirty space is contiguous across a number of buffers > > we should be able to send a *SINGLE* commit rpc to the server. That would > > greatly reduce system overhead on both the client and server when writing > > a large file over NFS. This would seem to be an almost free optimization > > that would mesh extremely well with the nfsrv_commit optimizations I'm > > making right now. > > > > At the moment I can saturate a 100BaseTX with NFS writes and get > > 10 MBytes/sec to the platter on the server, but the cpu required on both > > the client and server to do that is well over 60% of a Pentium III-450. > > I'd like to put in a vote to get these NFS write optimizations in the pipe > ASAP. The last hurdle (seemingly) on my big project at work that I've been > bugging so many of you about is the fact that FreeBSD NFS client writing to > Sun NFS server is just DOG slow. I did some pretty extensive testing on > this and couldn't come up with any client option twidding that made any > difference, except increasing wsize to 16k, which got me about 10%, but it > was still very slow. This is on a -current system from around the tenth of > August. > > Matt, thanks for all your hard work on this, and believe me when I say it > couldn't come at a better time. > > Doug > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-current" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message