Re: IDL and Beowolf ? [message #26103 is a reply to message #25991] |
Fri, 03 August 2001 10:06   |
Axel Schweiger
Messages: 11 Registered: November 1995
|
Junior Member |
|
|
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
On a slightly different note. My sense is that a great deal of progress
in
<br>terms of speed for big applications could be made by having a
translator/compiler
<br>for IDL. I know you can write code to run efficiently in IDL as
well but becomes
<br>very hard when you have a lot of branches in your code. Just
my 2 cents worth.
<p>Axel
<br>
<p>"Steve Smith" wrote:
<blockquote TYPE=CITE>On 31 Jul 2001 12:53:47 -0700, David Bruhwiler <bruhwile@txcorp.com>
wrote:
<br>>dmarshall@ivory.trentu.ca wrote in message news:
<br>>> I just finished reading The Do-It-Yourself Supercomputer by William
W.
<br>>> Hargrove, Forrest M. Hoffman and Thomas Sterling in this months
Scientific
<br>>> American
<br>>> <a href="http://www.sciam.com/2001/0801issue/0801hargrove.html">http://www.sciam.com/2001/0801issue/0801hargrove.html</a>
<br>>> And wondered if anyone had any thoughts/experience with running
IDL
<br>>> applications on such a cluster.
<br>>>
<br>>> Dave.
<br>>
<br>>Hi Dave,
<br>>
<br>>At Tech-X Corporation, we have prototyped the use of MPI (message
<br>>passing interface) with IDL and have succesfully tested this simple
<br>>prototype on a Beowolf style linux cluster. Tech-X will soon
receive
<br>>funding from the US Department of Energy for further proof-of-principle
<br>>development of this idea. This next round of development will
occur
<br>>over a six month period, beginning in September.
<br>>
<br>>Tech-X just issued a press release on this subject, and I just posted
<br>>that press release to this news group.
<br>>
<br>>Cheers,
<br>>David
<p>Ugh, I just read your press release, ex-squeeze me for asking to re-post
it :^)!
<p>--
<br>Steve S.
<p>steve@NOSPAMmailaps.org
<br>remove NOSPAM before replying</blockquote>
<pre>--
#----------------------------------------------------------- ------------------
# Axel Schweiger & amp;nbsp; & amp;nbsp; & amp;nbsp; & amp;nbsp; & amp;nbsp; & amp;nbsp; & amp;nbsp; Tel:206-543-1312
# University of Washington Fax:206-616-3142
# Polar Science Center/Applied Physics Laboratory
# 1013 NE 40th Street & ;nbsp; & ;nbsp; & ;nbsp; & ;nbsp; & ;nbsp; & ;nbsp;
# Seattle, Wa 98105 & nbsp; & nbsp; & nbsp; & nbsp; & nbsp; axel@apl.washington.edu
#----------------------------------------------------------- ------------------ </pre>
</html>
|
|
|