[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Ys14oIHL85d/T7s+@dhcp22.suse.cz>
Date: Tue, 12 Jul 2022 15:35:28 +0200
From: Michal Hocko <mhocko@...e.com>
To: Abel Wu <wuyun.abel@...edance.com>
Cc: Gang Li <ligang.bdlg@...edance.com>, akpm@...ux-foundation.org,
surenb@...gle.com, hca@...ux.ibm.com, gor@...ux.ibm.com,
agordeev@...ux.ibm.com, borntraeger@...ux.ibm.com,
svens@...ux.ibm.com, viro@...iv.linux.org.uk,
ebiederm@...ssion.com, keescook@...omium.org, rostedt@...dmis.org,
mingo@...hat.com, peterz@...radead.org, acme@...nel.org,
mark.rutland@....com, alexander.shishkin@...ux.intel.com,
jolsa@...nel.org, namhyung@...nel.org, david@...hat.com,
imbrenda@...ux.ibm.com, adobriyan@...il.com,
yang.yang29@....com.cn, brauner@...nel.org,
stephen.s.brennan@...cle.com, zhengqi.arch@...edance.com,
haolee.swjtu@...il.com, xu.xin16@....com.cn,
Liam.Howlett@...cle.com, ohoono.kwon@...sung.com,
peterx@...hat.com, arnd@...db.de, shy828301@...il.com,
alex.sierra@....com, xianting.tian@...ux.alibaba.com,
willy@...radead.org, ccross@...gle.com, vbabka@...e.cz,
sujiaxun@...ontech.com, sfr@...b.auug.org.au,
vasily.averin@...ux.dev, mgorman@...e.de, vvghjk1234@...il.com,
tglx@...utronix.de, luto@...nel.org, bigeasy@...utronix.de,
fenghua.yu@...el.com, linux-s390@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
linux-mm@...ck.org, linux-perf-users@...r.kernel.org,
hezhongkun.hzk@...edance.com
Subject: Re: [PATCH v2 0/5] mm, oom: Introduce per numa node oom for
CONSTRAINT_{MEMORY_POLICY,CPUSET}
On Tue 12-07-22 19:12:18, Abel Wu wrote:
[...]
> I was just going through the mail list and happen to see this. There
> is another usecase for us about per-numa memory usage.
>
> Say we have several important latency-critical services sitting inside
> different NUMA nodes without intersection. The need for memory of these
> LC services varies, so the free memory of each node is also different.
> Then we launch several background containers without cpuset constrains
> to eat the left resources. Now the problem is that there doesn't seem
> like a proper memory policy available to balance the usage between the
> nodes, which could lead to memory-heavy LC services suffer from high
> memory pressure and fails to meet the SLOs.
I do agree that cpusets would be rather clumsy if usable at all in a
scenario when you are trying to mix NUMA bound workloads with those
that do not have any NUMA proferences. Could you be more specific about
requirements here though?
Let's say you run those latency critical services with "simple" memory
policies and mix them with the other workload without any policies in
place so they compete over memory. It is not really clear to me how can
you achieve any reasonable QoS in such an environment. Your latency
critical servises will be more constrained than the non-critical ones
yet they are more demanding AFAIU.
--
Michal Hocko
SUSE Labs
Powered by blists - more mailing lists