public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Lars Marowsky-Bree <lmb@suse.de>
To: Arjan van de Ven <arjanv@redhat.com>
Cc: Daniel Phillips <phillips@istop.com>,
	sdake@mvista.com, David Teigland <teigland@redhat.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [ANNOUNCE] Minneapolis Cluster Summit, July 29-30
Date: Mon, 12 Jul 2004 12:05:47 +0200	[thread overview]
Message-ID: <20040712100547.GF3933@marowsky-bree.de> (raw)
In-Reply-To: <1089615523.2806.5.camel@laptop.fenrus.com>

On 2004-07-12T08:58:46,
   Arjan van de Ven <arjanv@redhat.com> said:

> Running realtime and mlocked (prealloced) is most certainly not
> sufficient for causes like this; any system call that internally
> allocates memory (even if it's just for allocating the kernel side of
> the filename you handle to open) can lead to this RT, mlocked process to
> cause VM writeout elsewhere. 

Of course; appropriate safety measures - like not doing any syscall
which could potentially block, or isolating them from the main task via
double-buffering childs - need to be done. (heartbeat does this in
fact.)

Again, if we have "many" in kernel users requiring high performance &
low-latency, running in the kernel may not be as bad, but I still don't
entirely like it.

But user-space can also manage just fine, and instead continuing the "we
need highperf, low-latency and non-blocking so it must be in the
kernel", we may want to consider how to have high-perf low-latency
kernel/user-space communication so that we can NOT move this into the
kernel.

Suffice to say that many user-space implementations exist which satisfy
these needs quite sufficiently; in the case of a CFS, this argument may
be different, but I'd like to see some hard data to back it up.

(On a practical note, a system which drops out of membership because
allocating a 256 byte buffer for a filename takes longer than the node
deadtime (due to high load) is reasonably unlikely to be a healthy
cluster member anyway and is on its road to eviction already.)

The main reason why I'd like to see cluster infrastructure in the kernel
is not technical, but because it increases the pressure on unification
so much that people might actually get their act together this time ;-)


Sincerely,
    Lars Marowsky-Brée <lmb@suse.de>

-- 
High Availability & Clustering	    \ ever tried. ever failed. no matter.
SUSE Labs, Research and Development | try again. fail again. fail better.
SUSE LINUX AG - A Novell company    \ 	-- Samuel Beckett


  reply	other threads:[~2004-07-12 10:06 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-07-05  6:09 [ANNOUNCE] Minneapolis Cluster Summit, July 29-30 Daniel Phillips
2004-07-05 15:09 ` Christoph Hellwig
2004-07-05 18:42   ` Daniel Phillips
2004-07-05 19:08     ` Chris Friesen
2004-07-05 20:29       ` Daniel Phillips
2004-07-07 22:55         ` Steven Dake
2004-07-08  1:30           ` Daniel Phillips
2004-07-05 19:12     ` Lars Marowsky-Bree
2004-07-05 20:27       ` Daniel Phillips
2004-07-06  7:34         ` Lars Marowsky-Bree
2004-07-06 21:34           ` Daniel Phillips
2004-07-07 18:16             ` Lars Marowsky-Bree
2004-07-08  1:14               ` Daniel Phillips
2004-07-08  9:10                 ` Lars Marowsky-Bree
2004-07-08 10:53                   ` David Teigland
2004-07-08 14:14                     ` Chris Friesen
2004-07-08 16:06                       ` David Teigland
2004-07-08 18:22                     ` Daniel Phillips
2004-07-08 19:41                       ` Steven Dake
2004-07-10  4:58                         ` David Teigland
2004-07-10  4:58                         ` Daniel Phillips
2004-07-10 17:59                           ` Steven Dake
2004-07-10 20:57                             ` Daniel Phillips
2004-07-10 23:24                               ` Steven Dake
2004-07-11 19:44                                 ` Daniel Phillips
2004-07-11 21:06                                   ` Lars Marowsky-Bree
2004-07-12  6:58                                     ` Arjan van de Ven
2004-07-12 10:05                                       ` Lars Marowsky-Bree [this message]
2004-07-12 10:11                                         ` Arjan van de Ven
2004-07-12 10:21                                           ` Lars Marowsky-Bree
2004-07-12 10:28                                             ` Arjan van de Ven
2004-07-12 11:50                                               ` Lars Marowsky-Bree
2004-07-12 12:01                                                 ` Arjan van de Ven
2004-07-12 13:13                                                   ` Lars Marowsky-Bree
2004-07-12 13:40                                                     ` Nick Piggin
2004-07-12 20:54                                                       ` Andrew Morton
2004-07-13  2:19                                                         ` Daniel Phillips
2004-07-13  2:31                                                           ` Nick Piggin
2004-07-27  3:31                                                             ` Daniel Phillips
2004-07-27  4:07                                                               ` Nick Piggin
2004-07-27  5:57                                                                 ` Daniel Phillips
2004-07-14 12:19                                                         ` Pavel Machek
2004-07-15  2:19                                                           ` Nick Piggin
2004-07-15 12:03                                                             ` Marcelo Tosatti
2004-07-14  8:32                                             ` Pavel Machek
2004-07-12  4:08                                   ` Steven Dake
2004-07-12  4:23                                     ` Daniel Phillips
2004-07-12 18:21                                       ` Steven Dake
2004-07-12 19:54                                         ` Daniel Phillips
2004-07-13 20:06                                         ` Pavel Machek
2004-07-12 10:14                     ` Lars Marowsky-Bree
     [not found] <fa.io9lp90.1c02foo@ifi.uio.no>
     [not found] ` <fa.go9f063.1i72joh@ifi.uio.no>
2004-07-06  6:39   ` Aneesh Kumar K.V
  -- strict thread matches above, loose matches on Subject: below --
2004-07-10 14:58 James Bottomley
2004-07-10 16:04 ` David Teigland
2004-07-10 16:26   ` James Bottomley

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=20040712100547.GF3933@marowsky-bree.de \
    --to=lmb@suse.de \
    --cc=arjanv@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=phillips@istop.com \
    --cc=sdake@mvista.com \
    --cc=teigland@redhat.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