Paper Info
Title
Workshop on relaxing synchronization for multicore and manycore scalability (RACES 2012)
Abstract
Massively-parallel systems are coming: core counts keep rising whether conventional cores as in multicore and manycore systems, or specialized cores as in GPUs. Conventional wisdom has been to utilize this parallelism by reducing synchronization to the minimum required to preserve determinism in particular, by eliminating data races. However, Amdahl's law implies that on highly-parallel systems even a small amount of synchronization that introduces serialization will limit scaling. Thus, we are forced to confront the trade-off between synchronization and the ability of an implementation to scale performance with the number of processors: synchronization inherently limits parallelism. This workshop focuses on harnessing parallelism by limiting synchronization, even to the point where programs will compute inconsistent or approximate rather than exact answers.
Year
DOI
Venue
2012
10.1145/2384716.2384779
SPLASH
Keywords
Field
DocType
highly-parallel system,massively-parallel system,manycore scalability,conventional core,data race,manycore system,small amount,core count,exact answer,harnessing parallelism,conventional wisdom,algorithms,concurrency,scalability,synchronization,consistency,performance,data structures
Serialization,Programming language,Computer science,Real-time computing,Multi-core processor,Distributed computing,Synchronization,Amdahl's law,Data synchronization,Parallel computing,Data parallelism,Synchronization (computer science),Scalability
Conference
Citations 
PageRank 
References 
0
0.34
0
Authors
5
Name
Order
Citations
PageRank
Andrew P. Black11566366.84
Theo D'Hondt261859.24
Doug Kimelman335135.19
Martin C. Rinard44739277.55
David Ungar51530328.37