caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: "Mikkel Fahnøe Jørgensen" <mikkel@dvide.com>
To: OCaml <caml-list@inria.fr>
Subject: More cores
Date: Fri, 19 Dec 2008 14:04:40 +0100	[thread overview]
Message-ID: <caee5ad80812190504p14b7bdf8n9e04be0168adc43e@mail.gmail.com> (raw)

Is it time to start rethinking concurrency in OCaml?

I have followed the argumentation of only using one native thread for
the OCaml runtime.
I can easily see how this can increase performance and simplify implementation.
I can also see that spawning new processes makes sense, so you get a
local heap for each task.

However, as we move forward it seems that we will get more than a few
cores on the same computational node according to the following
article:

Intel says to prepare for 'thousands of cores':
http://news.cnet.com/8301-13924_3-9981760-64.html?part=rss&subj=news&tag=2547-1_3-0-20

As I see it, it is not feasible to spawn a new process with a local
heap for each core, when the number of cores increases dramatically.

I am not sure that a parallel GC is a sufficient solution either due
to the high contention on memory, at least unless it provide some
additional core affinity features. I believe some level of compiler
support is needed in the not so distant future such that enough
primitives are available to build powerful multi-core aware libraries.
One approach could be micro heaps with core affinity and handle
mutable memory specially.


Regards,
Mikkel


             reply	other threads:[~2008-12-19 13:06 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-12-19 13:04 Mikkel Fahnøe Jørgensen [this message]
2008-12-19 14:04 ` [Caml-list] " Dario Teixeira
2008-12-19 15:06   ` Alexy Khrabrov
2008-12-19 15:54     ` The Axis of Eval (was: More cores) Dario Teixeira
2008-12-19 16:26       ` [Caml-list] " Paolo Donadeo
2008-12-19 17:01       ` Dario Teixeira
2008-12-19 18:01         ` Christophe Raffalli
2008-12-19 18:50     ` [Caml-list] More cores Ulf Wiger (TN/EAB)
2008-12-19 19:10   ` Richard Jones
2008-12-19 22:31   ` Jon Harrop
2008-12-19 22:36     ` Erik de Castro Lopo
2008-12-19 22:53       ` Jon Harrop
2008-12-22 17:00         ` [Caml-list] More Caml Jon Harrop
2008-12-22 21:44           ` Richard Jones
2008-12-23  6:07             ` Jon Harrop
2008-12-23  9:59               ` Jon Harrop
2008-12-23 15:32                 ` Ashish Agarwal
2008-12-23 17:33                   ` Jon Harrop
2008-12-24 13:12                 ` Mikkel Fahnøe Jørgensen
2008-12-24 16:47                   ` Jon Harrop
2008-12-23 10:04               ` Richard Jones
2008-12-23 10:38                 ` Jon Harrop
2008-12-23  9:43           ` Oliver Bandel
2008-12-23 11:53             ` Jon Harrop
2008-12-19 22:42     ` [Caml-list] More cores Richard Jones
2008-12-20 19:33     ` Mikkel Fahnøe Jørgensen
2008-12-20 19:41       ` Mikkel Fahnøe Jørgensen
2008-12-19 20:37 ` Oliver Bandel
2008-12-19 21:27   ` Richard Jones
2008-12-19 22:03     ` Hezekiah M. Carty
2008-12-19 22:47       ` Richard Jones
2008-12-19 23:00         ` Alexy Khrabrov
2008-12-19 23:56         ` prelude.ml as another standard extension to Pervasives? Alexy Khrabrov
2008-12-20  1:40           ` [Caml-list] " Mikkel Fahnøe Jørgensen
2008-12-20  4:50             ` Alexy Khrabrov
2008-12-20 10:53               ` Zheng Li
2008-12-20 12:37         ` [Caml-list] More cores Richard Jones

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=caee5ad80812190504p14b7bdf8n9e04be0168adc43e@mail.gmail.com \
    --to=mikkel@dvide.com \
    --cc=caml-list@inria.fr \
    /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).