If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(NBC News)   The Roadrunner supercomputer is being shut down. Super geniuses everywhere mourn, cancel their ACME catalog subscription   (m.nbcnews.com) divider line 9
    More: Cool, roadrunner, Super Genius, supercomputers, astronomical catalog, Los Alamos National Laboratory, states with nuclear weapons, superstar, Los Alamos  
•       •       •

2209 clicks; posted to Geek » on 31 Mar 2013 at 8:57 AM (1 year ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



Voting Results (Smartest)
View Voting Results: Smartest and Funniest


Archived thread
2013-03-31 10:05:54 PM
1 votes:
upload.wikimedia.org

What a Roadrunner may look like.
2013-03-31 10:56:29 AM
1 votes:
Well, it's good that they shut that thing down.  It was going faster by the hour.
2013-03-31 10:50:43 AM
1 votes:

buzzcut73: So, new Fark server coming soon, or is Red Tube gonna snap it up at the surplus auction?

/Honestly curious about what's going to happen to it


The article mentioned that they are going to dismantle it.  What they're going to do with the parts, who knows what.
Considering how bad the government takes security over at Los Alamos, I wouldn't be surprised if they sell the parts and some of them end up in China or a rogue nation.
2013-03-31 10:43:08 AM
1 votes:
So, new Fark server coming soon, or is Red Tube gonna snap it up at the surplus auction?

/Honestly curious about what's going to happen to it
2013-03-31 10:37:31 AM
1 votes:

wildcardjack: I've never played with supercomputer scale problems but I've done a course in FEA. To properly describe an object would take massive matrices and lots of trigonometry, but it was all interconnected. So I understand why you'd want to keep your data in touch with all the other data points in the matrix, sort of.

It's strange to think about distributed computing vs super computing. An outsider has a hard time wrapping one's mind around the difference, and quite frankly I'm not sure where the line is. I know on DC you do single thread tasks that are not interdependent, but where's the point of complexity that breaks between DC and SC? I guess that's a CompSci issue, I just used ALGOR for my mechanical engineering course.

/Back in the day they'd do FEA by hand
//Instead of a steno pool they'd have a floor of women doing math.
///Which is why I think you should be able to break anything down to DC.


balance of  process/thread execution time vs communications latency and/or overall communications bandwidth
2013-03-31 10:04:30 AM
1 votes:
Imagine a Beowolf cluster of those things...
2013-03-31 10:04:04 AM
1 votes:
I've never played with supercomputer scale problems but I've done a course in FEA. To properly describe an object would take massive matrices and lots of trigonometry, but it was all interconnected. So I understand why you'd want to keep your data in touch with all the other data points in the matrix, sort of.

It's strange to think about distributed computing vs super computing. An outsider has a hard time wrapping one's mind around the difference, and quite frankly I'm not sure where the line is. I know on DC you do single thread tasks that are not interdependent, but where's the point of complexity that breaks between DC and SC? I guess that's a CompSci issue, I just used ALGOR for my mechanical engineering course.

/Back in the day they'd do FEA by hand
//Instead of a steno pool they'd have a floor of women doing math.
///Which is why I think you should be able to break anything down to DC.
2013-03-31 09:18:32 AM
1 votes:
img.photobucket.com
2013-03-31 09:10:07 AM
1 votes:

Enormous-Schwanstucker: Beep beep


Meep meep.

/FTFY
 
Displayed 9 of 9 comments

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »






Report