From: ron minnich <rminnich@gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Subject: Re: [9fans] πp
Date: Fri, 15 Oct 2010 09:45:34 -0700 [thread overview]
Message-ID: <AANLkTin_H-OCy2OyZTAubm9L0R7i7=xLb4b9wut0wF23@mail.gmail.com> (raw)
In-Reply-To: <10af1064d3ac35a8d2f62214d5eec485@gmx.de>
2010/10/15 <cinap_lenrek@gmx.de>:
> i wonder if making 9p work better over high latency connections is
> even the right answer to the problem.
The reason I care is that the link from a CPU node to a file server on
blue gene is high latency. It might as well be cross-country, it's so
bad.
> would it not be cool to have a way to
> teleport/migrate your process to a cpu server close to the data?
If only ... but Blue Gene is very heavy! It's too hard to carry around
like that.
> i know, this is a crazy blue sky idea that has lots of problems on its
> own... but it poped up again when i read the "bring the computation
> to the data" point from the ospray talk.
it's a very attractive idea that has been tried extensively at
different times and places for several decades and in almost all
cases has never worked out.
Here's a streaming idea that worked out well. In 1994, Larry McVoy
while at SGI did something called "NFS bypass". Basically the code did
the standard NFS lookup/getattr cycle but when it came time to do the
read, what was returned was a TCP socket that in effect implemented
streamed data from server to client. Flow control became a TCP
problem, and SGI TCP did a good job with it. At that point the (fast
for the time) 800 mbit/second HIPPI channel ran at full rate, as
opposed to the much slower NFS file-oriented read. Simple
stream-oriented communications replaced all the games people play to
do prefetch, which in many cases is just making a packet-oriented
interface look like a stream anyway.
Look at it this way. Start rio under ratrace. Watch all the file IO.
You'd be hard pressed to find the ones that don't look like streams.
I also don't buy the "oh but this fails for a 2 GB file" argument.
Find me the 2 GB file you're using and then let's talk about not
streaming files that are too big. But if you are using a 2 GB *output*
file then you're almost always better off streaming that file out --
also verified on the LLNL Blue Gene where the current checkpointing of
32 TB of memory looks like a minor explosion, because the file system
protocol is such a bad fit to what is needed -- a stream from memory
to the server.
Streaming has been shown to work for HPC systems because most of our
applications stream data in our out. The transition from a stream to
the packet-oriented file IO protocol has never been comfortable. That
doesn't mean one only does streams, just that one should not make them
impossible.
ron
next prev parent reply other threads:[~2010-10-15 16:45 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-13 19:46 Eric Van Hensbergen
2010-10-13 22:13 ` David Leimbach
2010-10-13 22:36 ` roger peppe
2010-10-13 23:15 ` David Leimbach
2010-10-14 4:25 ` blstuart
2010-10-13 23:24 ` Venkatesh Srinivas
2010-10-14 21:12 ` Latchesar Ionkov
2010-10-14 21:48 ` erik quanstrom
2010-10-14 22:14 ` dorin bumbu
2010-10-15 14:07 ` erik quanstrom
2010-10-15 14:45 ` Russ Cox
2010-10-15 15:41 ` Latchesar Ionkov
2010-10-15 21:13 ` dorin bumbu
2010-10-15 22:13 ` erik quanstrom
2010-10-15 23:30 ` dorin bumbu
2010-10-15 23:45 ` cinap_lenrek
2010-10-15 23:54 ` dorin bumbu
2010-10-15 7:24 ` Sape Mullender
2010-10-15 10:41 ` hiro
2010-10-15 12:11 ` Jacob Todd
2010-10-15 13:06 ` Iruatã Souza
2010-10-15 13:23 ` Sape Mullender
2010-10-15 14:57 ` David Leimbach
2010-10-15 14:54 ` David Leimbach
2010-10-15 14:59 ` Francisco J Ballesteros
2010-10-15 16:19 ` cinap_lenrek
2010-10-15 16:28 ` Lyndon Nerenberg
2010-10-15 17:54 ` Nemo
2010-10-15 17:49 ` ron minnich
2010-10-15 18:56 ` erik quanstrom
2010-10-15 16:31 ` Latchesar Ionkov
2010-10-15 16:40 ` cinap_lenrek
2010-10-15 16:43 ` Latchesar Ionkov
2010-10-15 17:11 ` cinap_lenrek
2010-10-15 17:15 ` Latchesar Ionkov
2010-10-15 19:10 ` erik quanstrom
2010-10-15 19:16 ` Latchesar Ionkov
2010-10-15 21:43 ` Julius Schmidt
2010-10-15 22:02 ` David Leimbach
2010-10-15 22:05 ` John Floren
2010-10-15 16:45 ` ron minnich [this message]
2010-10-15 17:26 ` Eric Van Hensbergen
2010-10-15 17:45 ` Charles Forsyth
2010-10-15 17:45 ` ron minnich
2010-10-15 18:03 ` Bakul Shah
2010-10-15 18:45 ` David Leimbach
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='AANLkTin_H-OCy2OyZTAubm9L0R7i7=xLb4b9wut0wF23@mail.gmail.com' \
--to=rminnich@gmail.com \
--cc=9fans@9fans.net \
/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).