Deque-Free Work-Optimal Parallel STL Algorithms
Euro-Par '08 Proceedings of the 14th international Euro-Par conference on Parallel Processing
HyperSmooth: A System for Interactive Spatial Analysis Via Potential Maps
W2GIS '08 Proceedings of the 8th International Symposium on Web and Wireless Geographical Information Systems
Performance Evaluation of Work Stealing for Streaming Applications
OPODIS '09 Proceedings of the 13th International Conference on Principles of Distributed Systems
A mean field model of work stealing in large-scale systems
Proceedings of the ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Parallel arithmetic encryption for high-bandwidth communications on multicore/GPGPU platforms
Proceedings of the 4th International Workshop on Parallel and Symbolic Computation
Hardware/software support for adaptive work-stealing in on-chip multiprocessor
Journal of Systems Architecture: the EUROMICRO Journal
Output-sensitive decoding for redundant residue systems
Proceedings of the 2010 International Symposium on Symbolic and Algebraic Computation
Hi-index | 0.00 |
We study the problem of parallel stream computations on a multiprocessor architecture. Modelling the problem, we exhibit that any parallelisation introduces an arithmetic overhead related to intermediate copy operations. We provide lower bounds for the parallel stream computation on p processors of different speeds with two models, a strict model and a buffered model; to our knowledge, these are new results. We introduce a new parallel algorithm called processor-oblivious: it is based on the coupling of a fast sequential algorithm with a fine-grain parallel one that is scheduled by work-stealing. This algorithm is proved asymptotically optimal. We show that our algorithm has a good experimental behaviour.