From: Pavel Machek <pavel@ucw.cz>
To: Parth Shah <parth@linux.ibm.com>
Cc: peterz@infradead.org, mingo@redhat.com,
linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org,
patrick.bellasi@arm.com, dietmar.eggemann@arm.com,
daniel.lezcano@linaro.org, subhra.mazumdar@oracle.com
Subject: Re: [RFC v4 0/8] TurboSched: A scheduler for sustaining Turbo Frequencies for longer durations
Date: Wed, 31 Jul 2019 19:32:26 +0200 [thread overview]
Message-ID: <20190731173225.GB24222@amd> (raw)
In-Reply-To: <4fcd3488-6ba0-bc22-a08d-ceebbce1c120@linux.ibm.com>
[-- Attachment #1: Type: text/plain, Size: 2157 bytes --]
Hi!
> >> Abstract
> >> ========
> >>
> >> The modern servers allows multiple cores to run at range of frequencies
> >> higher than rated range of frequencies. But the power budget of the system
> >> inhibits sustaining these higher frequencies for longer durations.
> >
> > Thermal budget?
>
> Right, it is a good point, and there can be possibility of Thermal throttling
> which is not covered here.
> But the thermal throttling is less often seen in the servers than the throttling
> due to the Power budget constraints. Also one can change the power cap which leads
> to increase in the throttling and task packing can handle in such
> cases.
Ok. I thought you are doing this due to thermals. If I understand
things correctly, you can go over thermal limits for a few seconds
before the silicon heats up. What is the timescale for power budget?
> BTW, Task packing allows few more cores to remain idle for longer time, so
> shouldn't this decrease thermal throttles upto certain extent?
I guess so, yes.
> > >> These numbers are w.r.t. `turbo_bench.c` multi-threaded test benchmark
> >> which can create two kinds of tasks: CPU bound (High Utilization) and
> >> Jitters (Low Utilization). N in X-axis represents N-CPU bound and N-Jitter
> >> tasks spawned.
> >
> > Ok, so you have description how it causes 13% improvements. Do you also have metrics how
> > it harms performance.. how much delay is added to unimportant tasks etc...?
> >
>
> Yes, if we try to pack the tasks despite of no frequency throttling, we see a regression
> around 5%. For instance, in the synthetic benchmark I used to show performance benefit,
> for lower count of CPU intensive threads (N=2) there is -5% performance drop.
>
> Talking about the delay added to an unimportant tasks, the result can be lower throughput
> or higher latency for such tasks.
Thanks. I believe it would be good to mention disadvantages in the
documentation, too.
Best regards,
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
next prev parent reply other threads:[~2019-07-31 17:32 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-25 7:08 [RFC v4 0/8] TurboSched: A scheduler for sustaining Turbo Frequencies for longer durations Parth Shah
2019-07-25 7:08 ` [RFC v4 1/8] sched/core: Add manual jitter classification using sched_setattr syscall Parth Shah
2019-07-25 7:08 ` [RFC v4 2/8] sched: Introduce switch to enable TurboSched mode Parth Shah
2019-07-25 7:08 ` [RFC v4 3/8] sched/core: Update turbo_sched count only when required Parth Shah
2019-07-25 7:08 ` [RFC v4 4/8] sched/fair: Define core capacity to limit task packing Parth Shah
2019-07-25 7:08 ` [RFC v4 5/8] powerpc: Define Core Capacity for POWER systems Parth Shah
2019-07-25 7:08 ` [RFC v4 6/8] sched/fair: Tune task wake-up logic to pack jitter tasks Parth Shah
2019-07-25 7:08 ` [RFC v4 7/8] sched/fair: Bound non idle core search within LLC domain Parth Shah
2019-07-25 7:08 ` [RFC v4 8/8] powerpc: Set turbo domain to NUMA node for task packing Parth Shah
2019-07-28 13:31 ` [RFC v4 0/8] TurboSched: A scheduler for sustaining Turbo Frequencies for longer durations Pavel Machek
2019-07-31 16:39 ` Parth Shah
2019-07-31 17:32 ` Pavel Machek [this message]
2019-08-02 11:12 ` Parth Shah
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190731173225.GB24222@amd \
--to=pavel@ucw.cz \
--cc=daniel.lezcano@linaro.org \
--cc=dietmar.eggemann@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=parth@linux.ibm.com \
--cc=patrick.bellasi@arm.com \
--cc=peterz@infradead.org \
--cc=subhra.mazumdar@oracle.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).