From: "saeed bishara" Subject: Re: read-ahead in NFS server Date: Thu, 27 Dec 2007 17:00:12 +0200 Message-ID: References: <47730F2F.3080900@garzik.org> <47739288.7000308@garzik.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Cc: linux-kernel@vger.kernel.org, "NFS list" To: "Jeff Garzik" Return-path: Received: from wa-out-1112.google.com ([209.85.146.177]:21419 "EHLO wa-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751933AbXL0PAN (ORCPT ); Thu, 27 Dec 2007 10:00:13 -0500 Received: by wa-out-1112.google.com with SMTP id v27so4903454wah.23 for ; Thu, 27 Dec 2007 07:00:12 -0800 (PST) In-Reply-To: <47739288.7000308@garzik.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: > >> Are you using TCP? Are you using NFSv4, or an older version? > > I'm using NFSv3/UDP. > > IMO, you definitely want TCP and NFSv4. Much better network behavior, > with some of the silly UDP limits (plus greatly improved caching > behavior, due to v4 delegations). the clients of my system going to be embedded system with low performance cpus and I need UDP as it needs less cpu power. > > when I run local dd with bs=4K, I can see that the average IO size is > > more than 300KB. > > Read-ahead is easier in NFSv4, because the client probably has the file > delegated locally, and has far less need to constantly revalidate file > mapping(s). I'll check that. but what about the server side? why the issued IO's are only as twice as the size of the NFS requests?