From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:47843) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Z22OK-0005Mx-Q5 for qemu-devel@nongnu.org; Mon, 08 Jun 2015 15:07:30 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Z22OF-0007aB-Ss for qemu-devel@nongnu.org; Mon, 08 Jun 2015 15:07:24 -0400 Received: from e38.co.us.ibm.com ([32.97.110.159]:58930) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Z22OF-0007Yc-LQ for qemu-devel@nongnu.org; Mon, 08 Jun 2015 15:07:19 -0400 Received: from /spool/local by e38.co.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Mon, 8 Jun 2015 13:07:18 -0600 Received: from b03cxnp08025.gho.boulder.ibm.com (b03cxnp08025.gho.boulder.ibm.com [9.17.130.17]) by d03dlp01.boulder.ibm.com (Postfix) with ESMTP id B402B1FF004F for ; Mon, 8 Jun 2015 12:58:24 -0600 (MDT) Received: from d03av04.boulder.ibm.com (d03av04.boulder.ibm.com [9.17.195.170]) by b03cxnp08025.gho.boulder.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id t58J6d2g22020278 for ; Mon, 8 Jun 2015 12:06:39 -0700 Received: from d03av04.boulder.ibm.com (loopback [127.0.0.1]) by d03av04.boulder.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id t58J7E1i001989 for ; Mon, 8 Jun 2015 13:07:14 -0600 Message-ID: <5575E7E0.4050209@linux.vnet.ibm.com> Date: Mon, 08 Jun 2015 15:07:12 -0400 From: "Jason J. Herne" MIME-Version: 1.0 References: <1433267209-9882-1-git-send-email-jjherne@linux.vnet.ibm.com> <1433267209-9882-2-git-send-email-jjherne@linux.vnet.ibm.com> <87pp5drzrl.fsf@neno.neno> <556F412C.9090106@linux.vnet.ibm.com> <20150603180356.GE2129@work-vm> <556F4343.6070801@linux.vnet.ibm.com> In-Reply-To: <556F4343.6070801@linux.vnet.ibm.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v2 1/3] cpu: Provide vcpu throttling interface Reply-To: jjherne@linux.vnet.ibm.com List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Dr. David Alan Gilbert" Cc: amit.shah@redhat.com, borntraeger@de.ibm.com, quintela@redhat.com, qemu-devel@nongnu.org, afaerber@suse.de On 06/03/2015 02:11 PM, Jason J. Herne wrote: > On 06/03/2015 02:03 PM, Dr. David Alan Gilbert wrote: >> * Jason J. Herne (jjherne@linux.vnet.ibm.com) wrote: >>> On 06/03/2015 03:56 AM, Juan Quintela wrote: >>>> "Jason J. Herne" wrote: > ... >>>> We are checking for throotling on each cpu each 10ms. >>>> But on patch 2 we can see that we only change the throotling each >>>> time that we call migration_bitmap_sync(), that only happens each round >>>> through all the pages. Normally auto-converge only matters for machines >>>> with lots of memory, so this is going to happen each more than 10ms (we >>>> change it each 4 passes). You changed it to each 2 passes, and you add >>>> it a 0.2. I think that I would preffer to just have it each single >>>> pass, but add a 0.1 each pass? simpler and end result would be the >>>> same? >>>> >>>> >>> >>> Well, we certainly could make it run every pass but I think it would get >>> a little too aggressive then. The reason is, we do not increment the >>> throttle >>> rate by adding 0.2 each time. We increment it by multiplying the current >>> rate >>> by 2. So by doing that every pass we are doubling the exponential growth >>> rate. I will admit the numbers I chose are hardly scientific... I >>> chose them >>> because they seemed to provide a decent balance of "throttling >>> aggression" >>> in >>> my workloads. >> >> That's the advantage of making them parameters. > > I see your point. Expecting the user to configure these parameters > seems a bit much. But I guess, in theory, it is better to have the > ability to change them and not need it, than need it and not have it > right? > > So, as you stated earlier these should hook into MigrationParams > somehow? I'll admit this is the first I've seen this construct. If > this is the optimal location for the two controls (x-throttle-initial, > x-throttle-multiplier?) I can add them there. Will keep defaults of > 0.2 for initial and 2.0 for multiplier(is there a better name?)? > So I'm attempting add the initial throttle value and the multiplier to MigrationParameters and I've come across a problem. hmp_migrate_set_parameter assumes all parameters are ints. Apparently floating point is not allowed... void hmp_migrate_set_parameter(Monitor *mon, const QDict *qdict) { const char *param = qdict_get_str(qdict, "parameter"); int value = qdict_get_int(qdict, "value"); Also from hmp-commands.hx { .name = "migrate_set_parameter", .args_type = "parameter:s,value:i", .params = "parameter value", .help = "Set the parameter for migration", .mhandler.cmd = hmp_migrate_set_parameter, .command_completion = migrate_set_parameter_completion, }, I'm hoping someone already has an idea for dealing with this problem? If not, I suppose this is a good add-on for Dave's discussion on redesigning MigrationParameters. -- -- Jason J. Herne (jjherne@linux.vnet.ibm.com)