From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S966520Ab2EOU0f (ORCPT ); Tue, 15 May 2012 16:26:35 -0400 Received: from lennier.cc.vt.edu ([198.82.162.213]:51542 "EHLO lennier.cc.vt.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758169Ab2EOU0d (ORCPT ); Tue, 15 May 2012 16:26:33 -0400 X-Mailer: exmh version 2.8.0 04/21/2012 with nmh-1.4-dev To: Pantelis Antoniou Cc: Peter Zijlstra , Vincent Guittot , mou Chen , linux-kernel@vger.kernel.org, Ingo Molnar , torvalds@linux-foundation.org Subject: Re: Plumbers: Tweaking scheduler policy micro-conf RFP In-Reply-To: Your message of "Tue, 15 May 2012 14:35:53 +0300." <9093E8BA-80E4-4113-B036-0259E5FB44F1@gmail.com> From: valdis.kletnieks@vt.edu References: <833A8DB8-1AB4-45E4-8D44-14A0D782807D@gmail.com> <1337077714.27694.21.camel@twins> <9093E8BA-80E4-4113-B036-0259E5FB44F1@gmail.com> Mime-Version: 1.0 Content-Type: multipart/signed; boundary="==_Exmh_1337113581_2773P"; micalg=pgp-sha1; protocol="application/pgp-signature" Content-Transfer-Encoding: 7bit Date: Tue, 15 May 2012 16:26:22 -0400 Message-ID: <54348.1337113582@turing-police.cc.vt.edu> X-Mirapoint-Received-SPF: 209.85.161.180 mail-gg0-f180.google.com valdis@vt.edu 4 softfail X-Junkmail-Status: score=10/50, host=vivi.cc.vt.edu X-Junkmail-Signature-Raw: score=unknown, refid=str=0001.0A020201.4FB2BBF9.001D,ss=1,re=0.000,fgs=0, ip=0.0.0.0, so=2011-07-25 19:15:43, dmn=2011-05-27 18:58:46, mode=single engine X-Junkmail-IWF: false Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --==_Exmh_1337113581_2773P Content-Type: text/plain; charset=us-ascii On Tue, 15 May 2012 14:35:53 +0300, Pantelis Antoniou said: > Thermal management: How to distribute load to the processors in such > a way that the temperature of the die doesn't increase too much that > we have to either go to a lower OPP or shut down the core all-together. > This is in direct conflict with throughput since we'd have better performance > if we could keep the same warmed-up cpu going. It's not just "temperature of the die". When you have multiple aisles of 42U racks full of servers, you often hit "must keep average total BTU load per server below X" constraints. There's plenty of colo's that are only using 40% of their floor space due to cooling constraints (you may be able to get the power company to pull another megawatt of copper into the building, but then you need to find someplace to put another megawatt worth of cooling). --==_Exmh_1337113581_2773P Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) Comment: Exmh version 2.5 07/13/2001 iQIVAwUBT7K77QdmEQWDXROgAQLyow/8DjWVxCbuPMUeX4hu2yn7ZqfmF7bTBdH7 v96Xg2Oieg773RDKf4Si6IGtX3xKSkkauNaVDzoG1xCYb138OyN1zwNb2uFmlh6f JYBBwWsNrhidIzIN5SHz7tB4E/rxeGLdUkPBy7VMokYAJRFQHw4m2h4OVpENRtXO qpv5ObnI5B6N6IWYihRtoGFDTfBhYDV4ef2ofChy8BIBu3Yh4O65T/geJfmdHrSj sMxe81KnYI/ibSq5T+4+llq34HJ1ugMMZAS3TJpjth8Wm/zYpnJm+TK8OnmH3IBP o92wWF5tQ+hMRqYuEgFQujbCnf79k5ix37cWx7s7cJh08Lpxjl2d4EfmwDMzm6Dg p2xNUS0haHMPMeI2AbKI12D5qeolaGMKMXp/Yb5lwtO2RV+Q6eu1HTM0epj6x8Gj i2ejMYbiaIWU/HsHPqWP5TXOAs1HfNcBrTU6p8dVNRTLVrlARDRXUbdRxny8GPqz fEGBg8SmzPpzfQrsTL+fYXWsPFGCtNT1/7gcAnA2sQmcHUmazNj3VDhNhE2uSTb/ R1RASDu+gMROMM3eqd45KrPXqwFLDqmaLK/Qb+pRAEts54vaeR+1yLPCPqgl8Yvq MRBDn0CiiBC4MtpKCQpy4lz+P0weRG3V4VXA3zDYwx1i5NtvRFrfoIazYjD4Q9Ss l9Qt19hmWUI= =/CsH -----END PGP SIGNATURE----- --==_Exmh_1337113581_2773P--