From: Yang Shi <yang.shi@linux.alibaba.com>
To: Keith Busch <keith.busch@intel.com>, Dave Hansen <dave.hansen@intel.com>
Cc: Michal Hocko <mhocko@kernel.org>,
mgorman@techsingularity.net, riel@surriel.com,
hannes@cmpxchg.org, akpm@linux-foundation.org,
dan.j.williams@intel.com, fengguang.wu@intel.com,
fan.du@intel.com, ying.huang@intel.com, ziy@nvidia.com,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [v2 RFC PATCH 0/9] Another Approach to Use PMEM as NUMA Node
Date: Thu, 18 Apr 2019 12:23:41 -0700 [thread overview]
Message-ID: <8259dfd6-9044-b9f8-29b1-f427b4435eda@linux.alibaba.com> (raw)
In-Reply-To: <20190418181643.GB7659@localhost.localdomain>
On 4/18/19 11:16 AM, Keith Busch wrote:
> On Wed, Apr 17, 2019 at 10:13:44AM -0700, Dave Hansen wrote:
>> On 4/17/19 2:23 AM, Michal Hocko wrote:
>>> yes. This could be achieved by GFP_NOWAIT opportunistic allocation for
>>> the migration target. That should prevent from loops or artificial nodes
>>> exhausting quite naturaly AFAICS. Maybe we will need some tricks to
>>> raise the watermark but I am not convinced something like that is really
>>> necessary.
>> I don't think GFP_NOWAIT alone is good enough.
>>
>> Let's say we have a system full of clean page cache and only two nodes:
>> 0 and 1. GFP_NOWAIT will eventually kick off kswapd on both nodes.
>> Each kswapd will be migrating pages to the *other* node since each is in
>> the other's fallback path.
>>
>> I think what you're saying is that, eventually, the kswapds will see
>> allocation failures and stop migrating, providing hysteresis. This is
>> probably true.
>>
>> But, I'm more concerned about that window where the kswapds are throwing
>> pages at each other because they're effectively just wasting resources
>> in this window. I guess we should figure our how large this window is
>> and how fast (or if) the dampening occurs in practice.
> I'm still refining tests to help answer this and have some preliminary
> data. My test rig has CPU + memory Node 0, memory-only Node 1, and a
> fast swap device. The test has an application strict mbind more than
> the total memory to node 0, and forever writes random cachelines from
> per-cpu threads.
Thanks for the test. A follow-up question, how about the size for each
node? Is node 1 bigger than node 0? Since PMEM typically has larger
capacity, so I'm wondering whether the capacity may make things
different or not.
> I'm testing two memory pressure policies:
>
> Node 0 can migrate to Node 1, no cycles
> Node 0 and Node 1 migrate with each other (0 -> 1 -> 0 cycles)
>
> After the initial ramp up time, the second policy is ~7-10% slower than
> no cycles. There doesn't appear to be a temporary window dealing with
> bouncing pages: it's just a slower overall steady state. Looks like when
> migration fails and falls back to swap, the newly freed pages occasionaly
> get sniped by the other node, keeping the pressure up.
next prev parent reply other threads:[~2019-04-18 19:23 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-11 3:56 [v2 RFC PATCH 0/9] Another Approach to Use PMEM as NUMA Node Yang Shi
2019-04-11 3:56 ` [v2 PATCH 1/9] mm: define N_CPU_MEM node states Yang Shi
2019-04-11 3:56 ` [v2 PATCH 2/9] mm: page_alloc: make find_next_best_node find return cpuless node Yang Shi
2019-04-11 3:56 ` [v2 PATCH 3/9] mm: numa: promote pages to DRAM when it gets accessed twice Yang Shi
2019-04-11 3:56 ` [v2 PATCH 4/9] mm: migrate: make migrate_pages() return nr_succeeded Yang Shi
2019-04-11 3:56 ` [v2 PATCH 5/9] mm: vmscan: demote anon DRAM pages to PMEM node Yang Shi
2019-04-11 14:31 ` Dave Hansen
2019-04-15 22:10 ` Yang Shi
2019-04-15 22:14 ` Dave Hansen
2019-04-15 22:26 ` Yang Shi
2019-04-11 3:56 ` [v2 PATCH 6/9] mm: vmscan: don't demote for memcg reclaim Yang Shi
2019-04-11 3:56 ` [v2 PATCH 7/9] mm: vmscan: check if the demote target node is contended or not Yang Shi
2019-04-11 16:06 ` Dave Hansen
2019-04-15 22:06 ` Yang Shi
2019-04-15 22:13 ` Dave Hansen
2019-04-15 22:23 ` Yang Shi
2019-04-11 3:56 ` [v2 PATCH 8/9] mm: vmscan: add page demotion counter Yang Shi
2019-04-11 3:56 ` [v2 PATCH 9/9] mm: numa: add page promotion counter Yang Shi
2019-04-11 14:28 ` [v2 RFC PATCH 0/9] Another Approach to Use PMEM as NUMA Node Dave Hansen
2019-04-12 8:47 ` Michal Hocko
2019-04-16 0:09 ` Yang Shi
2019-04-16 7:47 ` Michal Hocko
2019-04-16 14:30 ` Dave Hansen
2019-04-16 14:39 ` Michal Hocko
2019-04-16 15:46 ` Dave Hansen
2019-04-16 18:34 ` Michal Hocko
2019-04-16 15:33 ` Zi Yan
2019-04-16 15:55 ` Dave Hansen
2019-04-16 16:12 ` Zi Yan
2019-04-16 19:19 ` Yang Shi
2019-04-16 21:22 ` Dave Hansen
2019-04-16 21:59 ` Yang Shi
2019-04-16 23:04 ` Dave Hansen
2019-04-16 23:17 ` Yang Shi
2019-04-17 15:13 ` Keith Busch
2019-04-17 9:23 ` Michal Hocko
2019-04-17 15:23 ` Keith Busch
2019-04-17 15:39 ` Michal Hocko
2019-04-17 15:37 ` Keith Busch
2019-04-17 16:39 ` Michal Hocko
2019-04-17 17:26 ` Yang Shi
2019-04-17 17:29 ` Keith Busch
2019-04-17 17:51 ` Michal Hocko
2019-04-18 16:24 ` Yang Shi
2019-04-17 17:13 ` Dave Hansen
2019-04-17 17:57 ` Michal Hocko
2019-04-18 18:16 ` Keith Busch
2019-04-18 19:23 ` Yang Shi [this message]
2019-04-18 21:07 ` Zi Yan
2019-04-16 23:18 ` Yang Shi
2019-04-17 9:17 ` Michal Hocko
2019-05-01 6:43 ` Fengguang Wu
2019-04-17 20:43 ` Yang Shi
2019-04-18 9:02 ` Michal Hocko
2019-05-01 5:20 ` Fengguang Wu
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=8259dfd6-9044-b9f8-29b1-f427b4435eda@linux.alibaba.com \
--to=yang.shi@linux.alibaba.com \
--cc=akpm@linux-foundation.org \
--cc=dan.j.williams@intel.com \
--cc=dave.hansen@intel.com \
--cc=fan.du@intel.com \
--cc=fengguang.wu@intel.com \
--cc=hannes@cmpxchg.org \
--cc=keith.busch@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=mhocko@kernel.org \
--cc=riel@surriel.com \
--cc=ying.huang@intel.com \
--cc=ziy@nvidia.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).