[aqm] avoiding the sqrt

Jeff Weeks <jweeks@sandvine.com> Mon, 10 August 2015 16:18 UTC

From: Jeff Weeks <jweeks@sandvine.com>
To: "aqm@ietf.org" <aqm@ietf.org>
Thread-Topic: avoiding the sqrt
Thread-Index: AdDThuCernObp5LLTKGgLqHu7Sbl8Q==
Date: Mon, 10 Aug 2015 16:18:26 +0000
Message-ID: <274D3A0FA900FD47AA6B56991AAA32FDC53BAA3F@wtl-exchp-1.sandvine.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/aqm/JYoQ6kJbkOngC6Gpgz0BtRyLQ9I>
Subject: [aqm] avoiding the sqrt
Precedence: list

The CoDel control_law is defined as 't + interval/sqrt(count)'

The sample implementation (http://queue.acm.org/appendices/codel.html) suggests that sqrt(count) can be calculated using only integer multiplication, but I'm wondering if it even needs to be calculated.

Would it not be more efficient to pre-calculate a set of 1/sqrt(count) for some range --  probably (1..interval), and for any count > interval, the next interval can just be set to t+1.

Is there any reason to not use this approach (assuming sufficient memory exists, of course)?

--Jeff

________________________________
/dev/jeff_weeks.x2936
Sandvine Incorporated

[aqm] avoiding the sqrt Jeff Weeks
Re: [aqm] avoiding the sqrt Mikael Abrahamsson
Re: [aqm] avoiding the sqrt Jonathan Morton
Re: [aqm] avoiding the sqrt Agarwal, Anil
Re: [aqm] avoiding the sqrt Jeff Weeks
Re: [aqm] avoiding the sqrt Andrew Mcgregor
Re: [aqm] avoiding the sqrt Dave Taht