From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Re: Latency difference between fifo and pfifo_fast Date: Wed, 7 Dec 2011 15:49:07 -0800 Message-ID: <20111207154907.462846cb@nehalam.linuxnetplumber.net> References: <20111207152709.37b5798d@nehalam.linuxnetplumber.net> <88e4825e-4efe-4163-bb24-299a20aab66d@jasiiieee> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: David Laight , netdev@vger.kernel.org, Rick Jones , Dave Taht , Eric Dumazet To: "John A. Sullivan III" Return-path: Received: from mail.vyatta.com ([76.74.103.46]:34450 "EHLO mail.vyatta.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751470Ab1LGXtK (ORCPT ); Wed, 7 Dec 2011 18:49:10 -0500 In-Reply-To: <88e4825e-4efe-4163-bb24-299a20aab66d@jasiiieee> Sender: netdev-owner@vger.kernel.org List-ID: = > > Is this a shared network? TOS won't matter if it is only your > > traffic. > > > > There are number of route metrics that you can tweak to that can > > reduce TCP slow > > start effects, like increasing the initial cwnd, etc. > > > It is a private network dedicated only to SAN traffic - a couple of SAN devices and some virtualization hosts - John Therefore unless your switch is shared, playing with queueing and TOS won't help reduce absolute latency. You maybe able to prioritize one host or SAN over another though.