Projects per year
This paper introduces hyperqueues, a programming abstraction that enables the construction of deterministic and scale-free pipeline parallel programs. Hyperqueues extend the concept of Cilk++ hyperobjects to provide thread-local views on a shared data structure. While hyperobjects are organized around private local views, hyperqueues require shared concurrent views on the underlying data structure. We define the semantics of hyperqueues and describe their implementation in a work-stealing scheduler. We demonstrate scalable performance on pipeline-parallel PARSEC benchmarks and find that hyperqueues provide comparable or up to 30% better performance than POSIX threads and Intel's Threading Building Blocks. The latter are highly tuned to the number of available processing cores, while programs using hyperqueues are scale-free.
|Title of host publication||SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis|
|Number of pages||12|
|Publication status||Published - Nov 2013|
|Event||SC13: 25th IEEE/ACM International Conference on High Performance Computing, Networking, Storage and Analysis - CO, Denver, United States|
Duration: 17 Nov 2014 → 21 Nov 2014
|Conference||SC13: 25th IEEE/ACM International Conference on High Performance Computing, Networking, Storage and Analysis|
|Period||17/11/2014 → 21/11/2014|
01/08/2012 → …
Invited Talk and membership in panel at the "European Initiative on Runtime Systems and Architecture Co-Design"
Hans Vandierendonck (Speaker)07 May 2015
Activity: Talk or presentation types › Invited talk