Return-Path: Received: from mail-bw0-f219.google.com ([209.85.218.219]:38496 "EHLO mail-bw0-f219.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753296AbZIBTKq convert rfc822-to-8bit (ORCPT ); Wed, 2 Sep 2009 15:10:46 -0400 Received: by bwz19 with SMTP id 19so961318bwz.37 for ; Wed, 02 Sep 2009 12:10:47 -0700 (PDT) In-Reply-To: <1f808b4a0909021135m49c2e60o6b305babcaed295c@mail.gmail.com> References: <20090902180841.GF946@proxime.net> <1f808b4a0909021135m49c2e60o6b305babcaed295c@mail.gmail.com> Date: Wed, 2 Sep 2009 15:10:47 -0400 Message-ID: Subject: Re: NFS for millions of files From: Aaron Wiebe To: Peter Chacko Cc: Jason Legate , linux-nfs@vger.kernel.org Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 On Wed, Sep 2, 2009 at 2:35 PM, Peter Chacko wrote: > Is this NFSv4 ? ?rsize and wsize > MTU size will cause fragmentation > and performance issues...Try making it around 4k .....You used ?1<<15 > fir your example. if you don't do writes....then this shouldn't > matter...and of course NFS is ?Nfs is not For Scalability.. You cannot > get the same performance on NFS as you would get for localFS...May be > you can try 10g....still there is TCP/UDP/IP stack overhead..... Fragmentation won't hurt you that much, and it doesn't even apply to opens, since those operations are significantly smaller. The performance gains higher in the stack from a larger rsize/wsize generally outweigh the savings in frame size optimization. Of course, jumbo frames are always a good idea, but again, doesn't help here. -Aaron