Title | ||
---|---|---|
Where should the threads go? Leveraging hierarchical data locality to solve the thread affinity dilemma |
Abstract | ||
---|---|---|
We are proposing a novel framework that ameliorates locality-aware parallel programming models, by defining a hierarchical data locality model extension. We also propose two hierarchical thread partitioning algorithms. These algorithms synthesize hierarchical thread placement layouts that targets minimizing the program's overall communication costs. We demonstrate the effectiveness of our approach using the NAS Parallel Benchmarks implemented in Unified Parallel C (UPC) using a modified Berkeley UPC Compiler and runtime system. We achieved performance gains of up to 88% in performance by applying the placement layouts our algorithms suggest. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/PADSW.2014.7097832 | ICPADS |
Keywords | Field | DocType |
nas parallel benchmark,communication cost minimization,data locality model extension,multi-threading,unified parallel c,hierarchical thread partitioning algorithm,thread affinity dilemma,berkeley upc compiler,many-cores,hierarchical locality,data handling,locality-aware parallel programming model,minimisation,runtime system,data locality,program compilers | Locality,Unified Parallel C,Computer science,Parallel computing,Compiler,Thread (computing),Processor affinity,Dilemma,Hierarchical database model,Distributed computing,Runtime system | Conference |
ISSN | Citations | PageRank |
1521-9097 | 0 | 0.34 |
References | Authors | |
9 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ahmad Anbar | 1 | 6 | 2.47 |
A. A. Badawy | 2 | 67 | 16.88 |
Olivier Serres | 3 | 65 | 7.52 |
tarek elghazawi | 4 | 697 | 84.30 |