From mboxrd@z Thu Jan 1 00:00:00 1970 From: jamal Subject: Re: [RFC][PATCH 1/3] NET_SCHED: PSPacer qdisc module Date: Wed, 21 Nov 2007 10:15:30 -0500 Message-ID: <1195658130.4437.17.camel@localhost> References: <20071121.191840.07671434.takano@axe-inc.co.jp> Reply-To: hadi@cyberus.ca Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, shemminger@linux-foundation.org, t.kudoh@aist.go.jp To: Ryousei Takano Return-path: Received: from py-out-1112.google.com ([64.233.166.182]:49136 "EHLO py-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755546AbXKUPPo (ORCPT ); Wed, 21 Nov 2007 10:15:44 -0500 Received: by py-out-1112.google.com with SMTP id u77so7253916pyb for ; Wed, 21 Nov 2007 07:15:35 -0800 (PST) In-Reply-To: <20071121.191840.07671434.takano@axe-inc.co.jp> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Wed, 2007-21-11 at 19:18 +0900, Ryousei Takano wrote: > This patch includes the PSPacer (Precise Software Pacer) qdisc > module, which achieves precise transmission bandwidth control. > You can find more information at the project web page > (http://www.gridmpi.org/gridtcp.jsp). Good stuff. I have not read your paper - There are NICs out there (chelsio comes to mind) which claim to do pacing and have shown impressive numbers with TCP. Is your approach similar? Are there patents involved by some of these hardware vendors? (It would not be suprising if they exist). The advantage with NICs is they have very good control of the timing (clock granularity being extremely important in cases like this) - what were your measurements based on i.e what clock source did you use on Linux? Also, the idea of using a PAUSE frame to add gaps is interesting, but you should note that in linux a qdisc may be attached to any network device and this for example maybe a PPP device etc. What would you use for gaps in that case? I apologize if the answers are in your papers - i just glossed over. cheers, jamal