From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-yh0-x229.google.com (mail-yh0-x229.google.com [IPv6:2607:f8b0:4002:c01::229]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 07CC121F19B for ; Fri, 3 Apr 2015 02:28:05 -0700 (PDT) Received: by yhme10 with SMTP id e10so3063241yhm.3 for ; Fri, 03 Apr 2015 02:28:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=3SSftFQI7snCwFwQNr1BACGQH3gR6m9hNx+de3tzv5M=; b=R2Ic2q6UB01MoA18jAal2c6G7nhsOmMlIh+oA/DFhwucXzfB66yvTd+jD8DZYlR94b S4K0hknp2G/zX2XUks2NJHAB3NHE9+S4zzcYu+GXYzPS4wW0gAZQgVkZN0U2rZ4A0soa dHV4vD8/JNcDgE1zjqLVgfTYwMEtEItHA0H/hWwLU2vJ28ZtSJpkdKPlZ6Sxkt8x79XI PhuBL9r2FsTZqoQ1pnrQ3o6myhSeGiKqmmb0syznNWBtJLKnYxb3XeFpiBVXOVxnCR8J VaFLtb0QA2VMQ8Tyqr4sHlXsS5KgFBl0H97BVYVkH+t2aKbrgsG8LoGoYszGu+YzCR6f UTYw== MIME-Version: 1.0 X-Received: by 10.52.65.233 with SMTP id a9mr915554vdt.42.1428053284624; Fri, 03 Apr 2015 02:28:04 -0700 (PDT) Received: by 10.52.240.145 with HTTP; Fri, 3 Apr 2015 02:28:04 -0700 (PDT) Received: by 10.52.240.145 with HTTP; Fri, 3 Apr 2015 02:28:04 -0700 (PDT) In-Reply-To: References: <551E364D.3070006@superduper.net> Date: Fri, 3 Apr 2015 12:28:04 +0300 Message-ID: From: Jonathan Morton To: David Lang Content-Type: multipart/alternative; boundary=20cf307f3b04c01fad0512ce9010 Cc: bloat Subject: Re: [Bloat] Computer generated congestion control 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: Fri, 03 Apr 2015 09:28:34 -0000 --20cf307f3b04c01fad0512ce9010 Content-Type: text/plain; charset=UTF-8 > > I'd like them to put some sane upper bound on the RTT - one compatible with satellite links, but which would avoid flooding unmanaged buffers to multi-minute delays. > The problem is that there aren't any numbers that meet these two criteria. > Even if you ignore 10G and faster interfaces, a 1Gb/s interface withsatellite sized latencies is a LOT of data, far more than is needed to flood a 'normal' link I very deliberately said "RTT", not "BDP". TCP stacks already track an estimate of RTT for various reasons, so in principle they could stop increasing the congestion window when that RTT reaches some critical value (1 second, say). The fact that they do not already do so is evidenced by the observations of multi-minute induced delays in certain circumstances. And this is not a complete solution by any means. Vegas proved that an altruistic limit on RTT by an endpoint, with no other measures within the network, leads to poor fairness between flows. But if the major OSes did that, more networks would be able to survive overload conditions while providing some usable service to their users. - Jonathan Morton --20cf307f3b04c01fad0512ce9010 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable

> > I'd like them to put some sane upper bound on = the RTT - one compatible with satellite links, but which would avoid floodi= ng unmanaged buffers to multi-minute delays.

> The problem is that there aren't any numbers that m= eet these two criteria.
> Even if you ignore 10G and faster interfaces, a 1Gb/s interface withsa= tellite sized latencies is a LOT of data, far more than is needed to flood = a 'normal' link

I very deliberately said "RTT", not "BDP"= ;.=C2=A0 TCP stacks already track an estimate of RTT for various reasons, s= o in principle they could stop increasing the congestion window when that R= TT reaches some critical value (1 second, say). The fact that they do not a= lready do so is evidenced by the observations of multi-minute induced delay= s in certain circumstances.

And this is not a complete solution by any means. Vegas prov= ed that an altruistic limit on RTT by an endpoint, with no other measures w= ithin the network, leads to poor fairness between flows. But if the major O= Ses did that, more networks would be able to survive overload conditions wh= ile providing some usable service to their users.

- Jonathan Morton

--20cf307f3b04c01fad0512ce9010--