Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).

×

+ - Relaxed accuracy doesn't imply catastrophic failure->

Submitted by QQBoss
QQBoss (2527196) writes "Coming from the mostly embedded world as I do, where a missed deadline or wrong value can mean destroyed hardware (or, more importantly, dead people), the thought of an inaccurate CPU is more than a little troubling. The authors of this paper from MIT understand that, and so they have developed mathematical frameworks to show you can rigorously limit the possibility of catastrophic failure.

Two years ago, loop perforation was shown to provide results faster and/or more efficiently than traditional CPUs by trading accuracy for performance, but the June paper presentation at the Association for Computing Machinery's Conference on Programming Language Design and Implementation should help put minds like mine at ease that we can have it faster and cheaper (which qualifies as better, without the whole death side effect problem)."

Link to Original Source
This discussion was created for logged-in users only, but now has been archived. No new comments can be posted.

Relaxed accuracy doesn't imply catastrophic failure

Comments Filter:

Theory is gray, but the golden tree of life is green. -- Goethe

Working...