From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-vw0-f43.google.com (mail-vw0-f43.google.com [209.85.212.43]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id D69D8200051 for ; Sun, 2 Oct 2011 09:51:04 -0700 (PDT) Received: by vws13 with SMTP id 13so3922338vws.16 for ; Sun, 02 Oct 2011 09:51:03 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; bh=cZpj/yIIZmhc+yo5saYtJSQptNsR/DbDL5yDuGUTVLw=; b=Kiwfd8ckFMJBQbI5BCzBgMr+Ds5HMiW7czvDeb+z6tWf9BoX3YUB7B4fHVhZXpOrR0 a5AGw6qLGRSdXMgCnG7UZVQDSxp5wvFhecJJ/UnYFTkM487kkLx8FLwlw60QlZ6BDN3O fWxfD7bfD3kt5xXCLFBQMAQ14aX2DpLST4gUY= MIME-Version: 1.0 Received: by 10.52.180.4 with SMTP id dk4mr14108786vdc.49.1317574263410; Sun, 02 Oct 2011 09:51:03 -0700 (PDT) Received: by 10.220.95.11 with HTTP; Sun, 2 Oct 2011 09:51:03 -0700 (PDT) Date: Sun, 2 Oct 2011 18:51:03 +0200 Message-ID: From: Nimrod Aviram To: bloat@lists.bufferbloat.net Content-Type: multipart/alternative; boundary=bcaec517ac6aef8eac04ae53a9e8 Subject: [Bloat] Active Queue Management for applicative layer X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 02 Oct 2011 16:51:05 -0000 --bcaec517ac6aef8eac04ae53a9e8 Content-Type: text/plain; charset=ISO-8859-1 Hi all, I'm a newbie, so my apologies if I'm directing this question to the wrong list or anything :-) I'm writing code for an applicative layer (scsi), transported inside tcp (which is overall called iScsi). This applicative layer is very prone to timeouts, specifically anything not acked (in a scsi way) after 3 seconds is aborted, which means any bandwidth we already spent on the request was a complete waste. I'd like to implement an Active Queue Management scheme, so that latency is no longer than a second, and throughput is as high as possible. Is there any standard code that already does this, or anything similar? I would really appreciate any pointers you can throw my way, I'm kind of "in the dark" here. many thanks, Nimrod --bcaec517ac6aef8eac04ae53a9e8 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Hi all,
I'm a newbie, so my apologies if I'm di= recting this question to the wrong list or anything :-)

I'm writ= ing code for an applicative layer (scsi), transported inside tcp (which is = overall called iScsi).
This applicative layer is very prone to timeouts, specifically anything not= acked (in a scsi way) after 3 seconds is aborted,
which means any bandw= idth we already spent on the request was a complete waste.
I'd like = to implement an Active Queue Management scheme, so that latency is no longe= r than a second,
and throughput is as high as possible.
Is there any standard code that a= lready does this, or anything similar?
I would really appreciate any poi= nters you can throw my way, I'm kind of "in the dark" here.
many thanks,
Nimrod

--bcaec517ac6aef8eac04ae53a9e8--