Comments on: MIT team looks at CFD on multicore chips and finds less is more http://insidehpc.com/2010/04/27/mit-team-looks-at-cfd-on-multicore-chips-and-finds-less-is-more/ HPC News Without the Noise for Supercomputing Professionals | insideHPC Sun, 19 May 2013 10:46:54 +0000 hourly 1 http://wordpress.org/?v=3.1.1 By: What to read at insideHPC this week | insideHPC.com http://insidehpc.com/2010/04/27/mit-team-looks-at-cfd-on-multicore-chips-and-finds-less-is-more/#comment-224061 What to read at insideHPC this week | insideHPC.com Fri, 30 Apr 2010 13:48:49 +0000 http://insidehpc.com/?p=11125#comment-224061 [...] MIT team looks at CFD on multicore chips and finds less is more [...] [...] MIT team looks at CFD on multicore chips and finds less is more [...]

]]> By: %page% http://insidehpc.com/2010/04/27/mit-team-looks-at-cfd-on-multicore-chips-and-finds-less-is-more/#comment-223919 %page% Thu, 29 Apr 2010 23:46:59 +0000 http://insidehpc.com/?p=11125#comment-223919 [...] MIT team looks at CFD on multicore chips and finds less is more | insideHPC.com [...] [...] MIT team looks at CFD on multicore chips and finds less is more | insideHPC.com [...]

]]> By: John West http://insidehpc.com/2010/04/27/mit-team-looks-at-cfd-on-multicore-chips-and-finds-less-is-more/#comment-223669 John West Wed, 28 Apr 2010 19:03:34 +0000 http://insidehpc.com/?p=11125#comment-223669 Brian - thanks. I tried to word it such that I didn't claim it as original content for DD, because they have a bad habit of running press releases without noting them as such. I appreciate you keeping us honest. Brian – thanks. I tried to word it such that I didn’t claim it as original content for DD, because they have a bad habit of running press releases without noting them as such. I appreciate you keeping us honest.

]]> By: Brian http://insidehpc.com/2010/04/27/mit-team-looks-at-cfd-on-multicore-chips-and-finds-less-is-more/#comment-223644 Brian Wed, 28 Apr 2010 16:12:48 +0000 http://insidehpc.com/?p=11125#comment-223644 I'll just point out that this is not an original article written by Dr. Dobb's. Rather, it is a press release issued by MIT (see http://web.mit.edu/newsoffice/2010/multicore-0426.html). The author, Larry Hardesty, works for the MIT news office, though his affiliation isn't noted in the article published at DD. I’ll just point out that this is not an original article written by Dr. Dobb’s. Rather, it is a press release issued by MIT (see http://web.mit.edu/newsoffice/2010/multicore-0426.html). The author, Larry Hardesty, works for the MIT news office, though his affiliation isn’t noted in the article published at DD.

]]> By: John West http://insidehpc.com/2010/04/27/mit-team-looks-at-cfd-on-multicore-chips-and-finds-less-is-more/#comment-223622 John West Wed, 28 Apr 2010 14:16:42 +0000 http://insidehpc.com/?p=11125#comment-223622 Yes, Jeff, you are 100% right. I covered it mostly because SO MANY other people did. I felt like the story needed to be acknowledged here, and I did use it as a chance to get on my own soap box re SMP vs multicore. Yes, Jeff, you are 100% right. I covered it mostly because SO MANY other people did. I felt like the story needed to be acknowledged here, and I did use it as a chance to get on my own soap box re SMP vs multicore.

]]> By: Jeff Squyres http://insidehpc.com/2010/04/27/mit-team-looks-at-cfd-on-multicore-chips-and-finds-less-is-more/#comment-223454 Jeff Squyres Tue, 27 Apr 2010 23:08:50 +0000 http://insidehpc.com/?p=11125#comment-223454 I find it very hard to believe that so many people picked up on this story. At least John correctly identifies that THIS TECHNIQUE IS NOT NEW. Not even close. It actually offends me a bit that this "technique" is being paraded around as a new and/or multicore-specific technique (Dr. Dobbs, HPC Wire, etc.). It's neither. Work sharing is so not new that it was old when I was a grad student (and *that* was a long time ago). /me gets off my soap box... I find it very hard to believe that so many people picked up on this story. At least John correctly identifies that THIS TECHNIQUE IS NOT NEW. Not even close. It actually offends me a bit that this “technique” is being paraded around as a new and/or multicore-specific technique (Dr. Dobbs, HPC Wire, etc.). It’s neither.

Work sharing is so not new that it was old when I was a grad student (and *that* was a long time ago).

/me gets off my soap box…

]]>