Re: [PATCH 0/8][for -mm] mem_notify v6

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: KOSAKI Motohiro <kosaki.motohiro@...>
Cc: <linux-mm@...>, <linux-kernel@...>, <marcelo@...>, <daniel.spang@...>, <riel@...>, <akpm@...>, <alan@...>, <linux-fsdevel@...>, <pavel@...>, <a1426z@...>, <jonathan@...>, <zlynx@...>
Date: Tuesday, February 19, 2008 - 11:00 am

Kosaki-san wrote:

You're most welcome.  The pleasure is all mine.


There is not much my customers HPC jobs can do with notification before
swap.  Their jobs either have the main memory they need to perform the
requested calculations with the desired performance, or their job is
useless and should be killed.  Unlike the applications you describe,
my customers jobs have no way, once running, to adapt to less memory.
They can only adapt to less memory by being restarted with a different
set of resource requests to the job scheduler (the application that
manages job requests, assigns them CPU, memory and other resources,
and monitors, starts, stops and pauses jobs.)

The primary difficulty my HPC customers have is killing such jobs fast
enough, before a bad job (one that attempts to use more memory than it
signed up for) can harm the performance of other users and the rest of
the system.

I don't mind if a pages are slowly or occassionally written to swap;
but as soon as the task wants to reclaim big chunks of memory by
writing thousands of pages at once to swap, it must die, and die
before it can queue more than a handful of those pages to the swapper.


Yes - understood and agreed - as I guessed, cpusets are not configured
in embedded systems.


Yes - I agree that my ideas were quite different.  Please don't
hesitate to reject every one of them, like a Samurai slicing through
air with his favorite sword <grin>.


For your work, yes that hook is too late.  Agreed.

Depending on what we're trying to do:
 1) warn applications of swap coming soon (your case),
 2) show how close we are to swapping,
 3) show how much swap has happened already,
 4) kill instantly if try to swap (my hpc case),
 5) measure file i/o caused by memory pressure, or
 6) perhaps other goals,
we will need to hook different places in the kernel.

It may well be that your hooks for embedded are simply in different
places than my hooks for HPC.  If so, that's fine.

I look forward to your further thoughts.

-- 
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj@sgi.com> 1.940.382.4214
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Sat Feb 9, 11:19 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Tom May, (Tue Apr 1, 7:35 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Wed Apr 2, 3:31 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Tom May, (Mon Apr 14, 8:16 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Thu Apr 17, 5:30 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Tom May, (Thu Apr 17, 3:23 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Daniel Spång, (Wed Apr 23, 4:27 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Tom May, (Wed Apr 30, 10:07 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Thu May 1, 11:06 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Tom May, (Fri May 2, 6:21 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Sat May 3, 8:26 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Tom May, (Tue May 6, 1:22 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Fri Apr 18, 6:07 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Tom May, (Mon Apr 21, 4:32 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Tue Apr 15, 10:30 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Tom May, (Wed Apr 2, 1:45 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Paul Jackson, (Sun Feb 17, 10:49 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Tue Feb 19, 3:36 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Paul Jackson, (Tue Feb 19, 11:00 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Pavel Machek, (Tue Feb 19, 6:28 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Rik van Riel, (Tue Feb 19, 10:07 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Paul Jackson, (Wed Feb 20, 12:36 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Tue Feb 19, 10:48 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Paul Jackson, (Wed Feb 20, 12:57 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Wed Feb 20, 1:21 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Paul Jackson, (Tue Feb 19, 9:54 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Rik van Riel, (Tue Feb 19, 3:02 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Paul Jackson, (Tue Feb 19, 4:18 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Paul Jackson, (Tue Feb 19, 4:43 pm)
[PATCH 0/8][for -mm] mem_notify v6, Jonathan Corbet, (Mon Feb 11, 11:36 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Mon Feb 11, 11:46 am)
Re: [PATCH 0/8][for -mm] mem_notify v6, Jon Masters, (Sat Feb 9, 12:02 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Sat Feb 9, 12:33 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, Rik van Riel, (Sat Feb 9, 12:43 pm)
Re: [PATCH 0/8][for -mm] mem_notify v6, KOSAKI Motohiro, (Sat Feb 9, 12:49 pm)