From: Rik van Riel <riel@redhat.com>
To: Mel Gorman <mgorman@suse.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
Andrea Arcangeli <aarcange@redhat.com>,
Ingo Molnar <mingo@kernel.org>,
Johannes Weiner <hannes@cmpxchg.org>,
Hugh Dickins <hughd@google.com>,
Thomas Gleixner <tglx@linutronix.de>,
Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>,
Linux-MM <linux-mm@kvack.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 16/19] mm: numa: Add pte updates, hinting and migration stats
Date: Tue, 06 Nov 2012 14:55:06 -0500 [thread overview]
Message-ID: <50996B1A.7040601@redhat.com> (raw)
In-Reply-To: <1352193295-26815-17-git-send-email-mgorman@suse.de>
On 11/06/2012 04:14 AM, Mel Gorman wrote:
> It is tricky to quantify the basic cost of automatic NUMA placement in a
> meaningful manner. This patch adds some vmstats that can be used as part
> of a basic costing model.
>
> u = basic unit = sizeof(void *)
> Ca = cost of struct page access = sizeof(struct page) / u
> Cpte = Cost PTE access = Ca
> Cupdate = Cost PTE update = (2 * Cpte) + (2 * Wlock)
> where Cpte is incurred twice for a read and a write and Wlock
> is a constant representing the cost of taking or releasing a
> lock
> Cnumahint = Cost of a minor page fault = some high constant e.g. 1000
> Cpagerw = Cost to read or write a full page = Ca + PAGE_SIZE/u
> Ci = Cost of page isolation = Ca + Wi
> where Wi is a constant that should reflect the approximate cost
> of the locking operation
> Cpagecopy = Cpagerw + (Cpagerw * Wnuma) + Ci + (Ci * Wnuma)
> where Wnuma is the approximate NUMA factor. 1 is local. 1.2
> would imply that remote accesses are 20% more expensive
>
> Balancing cost = Cpte * numa_pte_updates +
> Cnumahint * numa_hint_faults +
> Ci * numa_pages_migrated +
> Cpagecopy * numa_pages_migrated
>
> Note that numa_pages_migrated is used as a measure of how many pages
> were isolated even though it would miss pages that failed to migrate. A
> vmstat counter could have been added for it but the isolation cost is
> pretty marginal in comparison to the overall cost so it seemed overkill.
>
> The ideal way to measure automatic placement benefit would be to count
> the number of remote accesses versus local accesses and do something like
>
> benefit = (remote_accesses_before - remove_access_after) * Wnuma
>
> but the information is not readily available. As a workload converges, the
> expection would be that the number of remote numa hints would reduce to 0.
>
> convergence = numa_hint_faults_local / numa_hint_faults
> where this is measured for the last N number of
> numa hints recorded. When the workload is fully
> converged the value is 1.
>
> This can measure if the placement policy is converging and how fast it is
> doing it.
>
> Signed-off-by: Mel Gorman <mgorman@suse.de>
I'm skipping the ACKing of the policy patches, which
appear to be meant to be placeholders for a "real"
policy. However, you have a few more mechanism patches
left in the series, which would be required regardless
of what policy gets merged, so ...
Reviewed-by: Rik van Riel <riel@redhat.com>
next prev parent reply other threads:[~2012-11-06 19:52 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-06 9:14 [RFC PATCH 00/19] Foundation for automatic NUMA balancing Mel Gorman
2012-11-06 9:14 ` [PATCH 01/19] mm: compaction: Move migration fail/success stats to migrate.c Mel Gorman
2012-11-06 17:32 ` Rik van Riel
2012-11-06 9:14 ` [PATCH 02/19] mm: migrate: Add a tracepoint for migrate_pages Mel Gorman
2012-11-06 17:33 ` Rik van Riel
2012-11-06 9:14 ` [PATCH 03/19] mm: compaction: Add scanned and isolated counters for compaction Mel Gorman
2012-11-06 17:35 ` Rik van Riel
2012-11-06 9:14 ` [PATCH 04/19] mm: numa: define _PAGE_NUMA Mel Gorman
2012-11-06 18:35 ` Rik van Riel
2012-11-06 9:14 ` [PATCH 05/19] mm: numa: pte_numa() and pmd_numa() Mel Gorman
2012-11-13 9:54 ` Ingo Molnar
2012-11-13 11:24 ` Mel Gorman
2012-11-06 9:14 ` [PATCH 06/19] mm: numa: teach gup_fast about pmd_numa Mel Gorman
2012-11-13 10:07 ` Ingo Molnar
2012-11-13 11:37 ` Mel Gorman
2012-11-13 13:51 ` Ingo Molnar
2012-11-06 9:14 ` [PATCH 07/19] mm: numa: split_huge_page: transfer the NUMA type from the pmd to the pte Mel Gorman
2012-11-06 9:14 ` [PATCH 08/19] mm: numa: Create basic numa page hinting infrastructure Mel Gorman
2012-11-06 18:58 ` Rik van Riel
2012-11-07 10:38 ` Mel Gorman
2012-11-07 10:48 ` Rik van Riel
2012-11-07 11:00 ` Mel Gorman
2012-11-13 10:21 ` Ingo Molnar
2012-11-13 11:50 ` Mel Gorman
2012-11-13 13:49 ` Ingo Molnar
2012-11-13 14:26 ` Mel Gorman
2012-11-06 9:14 ` [PATCH 09/19] mm: mempolicy: Make MPOL_LOCAL a real policy Mel Gorman
2012-11-06 9:14 ` [PATCH 10/19] mm: mempolicy: Add MPOL_MF_NOOP Mel Gorman
2012-11-06 9:14 ` [PATCH 11/19] mm: mempolicy: Check for misplaced page Mel Gorman
2012-11-06 9:14 ` [PATCH 12/19] mm: migrate: Introduce migrate_misplaced_page() Mel Gorman
2012-11-06 19:10 ` Rik van Riel
2012-11-13 9:36 ` Ingo Molnar
2012-11-13 11:43 ` Ingo Molnar
2012-11-13 11:56 ` Mel Gorman
2012-11-13 14:49 ` Rik van Riel
2012-11-06 9:14 ` [PATCH 13/19] mm: mempolicy: Use _PAGE_NUMA to migrate pages Mel Gorman
2012-11-06 19:18 ` Rik van Riel
2012-11-07 12:32 ` Mel Gorman
2012-11-06 9:14 ` [PATCH 14/19] mm: mempolicy: Add MPOL_MF_LAZY Mel Gorman
2012-11-06 19:19 ` Rik van Riel
2012-11-13 10:25 ` Ingo Molnar
2012-11-13 12:02 ` Mel Gorman
2012-11-06 9:14 ` [PATCH 15/19] mm: numa: Add fault driven placement and migration Mel Gorman
2012-11-06 19:41 ` Rik van Riel
2012-11-07 10:49 ` Mel Gorman
2012-11-07 11:46 ` Rik van Riel
2012-11-13 10:45 ` Ingo Molnar
2012-11-13 12:09 ` Mel Gorman
2012-11-13 13:39 ` Ingo Molnar
2012-11-06 9:14 ` [PATCH 16/19] mm: numa: Add pte updates, hinting and migration stats Mel Gorman
2012-11-06 19:55 ` Rik van Riel [this message]
2012-11-07 10:57 ` Mel Gorman
2012-11-07 11:47 ` Rik van Riel
2012-11-06 9:14 ` [PATCH 17/19] mm: numa: Migrate on reference policy Mel Gorman
2012-11-07 11:56 ` Rik van Riel
2012-11-06 9:14 ` [PATCH 18/19] mm: sched: numa: Implement constant, per task Working Set Sampling (WSS) rate Mel Gorman
2012-11-06 19:55 ` Rik van Riel
2012-11-06 9:14 ` [PATCH 19/19] mm: sched: numa: Implement slow start for working set sampling Mel Gorman
2012-11-06 19:56 ` Rik van Riel
2012-11-07 9:27 ` [RFC PATCH 00/19] Foundation for automatic NUMA balancing Zhouping Liu
2012-11-07 15:25 ` Mel Gorman
2012-11-08 6:37 ` Zhouping Liu
[not found] ` <CAOHXNFG=T63dmc3smkJ2juE7HpxTv6qbavBXycRsXiLBzAwMGw@mail.gmail.com>
2012-11-08 7:03 ` Zhouping Liu
2012-11-09 14:42 ` Andrea Arcangeli
2012-11-09 16:12 ` Mel Gorman
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=50996B1A.7040601@redhat.com \
--to=riel@redhat.com \
--cc=a.p.zijlstra@chello.nl \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mingo@kernel.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
/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).