From: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
To: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Arjan van de Ven <arjan@infradead.org>,
Linux Kernel <linux-kernel@vger.kernel.org>,
venkatesh.pallipadi@intel.com, suresh.b.siddha@intel.com,
Michael Neuling <mikey@neuling.org>,
"Amit K. Arora" <aarora@linux.vnet.ibm.com>
Subject: Re: [RFC PATCH v1 0/3] Scaled statistics using APERF/MPERF in x86
Date: Tue, 27 May 2008 18:49:26 +0530 [thread overview]
Message-ID: <20080527131926.GB5181@dirshya.in.ibm.com> (raw)
In-Reply-To: <483C059B.9040301@linux.vnet.ibm.com>
[sniped]
> >> 3. How do I answer the following problem
> >>
> >> My CPU utilization is 50% at all frequencies (since utilization is
> >> time based), does it mean that frequency scaling does not impact my
> >> workload?
> >
> > without knowing anything else than this, then yes that would be a
> > logical conclusion: the most likely cause would be because your cpu is
> > memory bound. In fact, you could then scale down your cpu
> > frequency/voltage to be lower, and save some power without losing
> > performance.
> > It's a weird workload though, its probably a time based thing where you
> > alternate between idle and fully memory bound loads.
> >
> > (which is another case where your patches would then expose idle time
> > even though your cpu is fully utilized for the 50% of the time it's
> > running)
>
> We expect the end user to see 50% as scaled utilization and 100% as normal
> utilization. We don't intend to remove tsk->utime and tsk->stime. Our patches
> intend to provide the data and not impose what control action should be taken.
Hi Arjan,
As Balbir mentioned, we are not changing the idle time calculations.
The meaning of current utime and stime are preserved and they are
relative to current CPU capacity.
We are just adding a new metric (which already exist in taskstats) to
provide more utilisation data for higher level management software to
take decisions.
At any time we will have both the traditional utilisation value
relative to current CPU capacity, and scaled utilisation that is
relative to maximum CPU capacity.
--Vaidy
next prev parent reply other threads:[~2008-05-27 13:18 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-26 14:31 Vaidyanathan Srinivasan
2008-05-26 14:31 ` [RFC PATCH v1 1/3] General framework for APERF/MPERF access and accounting Vaidyanathan Srinivasan
2008-05-26 18:11 ` Balbir Singh
2008-05-27 14:54 ` Vaidyanathan Srinivasan
2008-05-26 14:31 ` [RFC PATCH v1 2/3] Make calls to account_scaled_stats Vaidyanathan Srinivasan
2008-05-26 18:18 ` Balbir Singh
2008-05-27 15:02 ` Vaidyanathan Srinivasan
2008-05-29 15:18 ` Michael Neuling
2008-05-29 18:23 ` Vaidyanathan Srinivasan
2008-05-26 14:31 ` [RFC PATCH v1 3/3] Print scaled utime and stime in getdelays Vaidyanathan Srinivasan
2008-05-26 15:50 ` [RFC PATCH v1 0/3] Scaled statistics using APERF/MPERF in x86 Arjan van de Ven
2008-05-26 17:24 ` Balbir Singh
2008-05-26 18:00 ` Arjan van de Ven
2008-05-26 18:36 ` Balbir Singh
2008-05-26 18:51 ` Arjan van de Ven
2008-05-27 12:59 ` Balbir Singh
2008-05-27 13:19 ` Vaidyanathan Srinivasan [this message]
2008-05-27 14:15 ` Arjan van de Ven
2008-05-27 15:27 ` Vaidyanathan Srinivasan
2008-05-31 21:27 ` Pavel Machek
2008-06-02 17:54 ` Vaidyanathan Srinivasan
2008-06-03 2:20 ` Arjan van de Ven
2008-05-27 13:29 ` Vaidyanathan Srinivasan
2008-05-27 14:19 ` Arjan van de Ven
2008-05-27 15:20 ` Vaidyanathan Srinivasan
2008-05-27 14:04 ` Vaidyanathan Srinivasan
2008-05-27 16:40 ` Arjan van de Ven
2008-05-27 18:26 ` Vaidyanathan Srinivasan
2008-05-31 21:17 ` Pavel Machek
2008-05-31 21:13 ` Pavel Machek
2008-06-02 6:08 ` Balbir Singh
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=20080527131926.GB5181@dirshya.in.ibm.com \
--to=svaidy@linux.vnet.ibm.com \
--cc=aarora@linux.vnet.ibm.com \
--cc=arjan@infradead.org \
--cc=balbir@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mikey@neuling.org \
--cc=suresh.b.siddha@intel.com \
--cc=venkatesh.pallipadi@intel.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
all inboxes | Powered by JetHome®