Home → Magazine Archive → November 2014 (Vol. 57, No. 11) → Researchers Simplify Parallel Programming → Abstract

Researchers Simplify Parallel Programming

By Gary Anthes

Communications of the ACM, Vol. 57 No. 11, Pages 13-15
10.1145/2667109

[article image]


Using statistical models, computer scientists have shown that certain kinds of parallel computation are not as difficult as previously thought. Researchers from the Massachusetts Institute of Technology (MIT), Microsoft Research, and the Israel Institute of Technology (Technion) showed that for a large class of non-blocking parallel programs, lock-free algorithms will perform fully as well as the more complex wait-free algorithms.

"What we have shown is that you really don't need to design these wait-free algorithms," said Nir Shavit, a computer science professor at MIT. "Most methods that are lock-free are also wait-free."

0 Comments

No entries found

Read CACM in a mobile app!
ACM Logo
  • ACM CACM apps for iPad, iPhone, and Android
  • ACM Digital Library app for iOS, Android, and Windows
  • Download free and sign in with ACM Web Account
Find the app for your mobile device
ACM DL Logo