Abstract | ||
---|---|---|
We propose a new software framework, named \"decoupling architecture\", for all-to-all computation. In this framework, the user's kernel code and well-tuned parallel control code are clearly separated to allows users to move smoothly from running sequential programs on a single-node server to fully utilizing a powerful machine, such as the K computer. In an evaluation of our prototype, the proposed decoupling architecture introduced some overhead. However, we show that the overhead can be reduced. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1145/2642769.2642801 | EuroMPI/ASIA |
Keywords | Field | DocType |
domain-specific architectures,big data,parallel programming,x-ray free electron laser,input/output,mpi | Kernel (linear algebra),Architecture,Computer science,Decoupling (cosmology),Parallel computing,Big data,Software framework,Computation | Conference |
Citations | PageRank | References |
0 | 0.34 | 3 |
Authors | ||
12 |
Name | Order | Citations | PageRank |
---|---|---|---|
Atsushi Hori | 1 | 50 | 5.50 |
Kazumi Yoshinaga | 2 | 6 | 2.68 |
Atsushi Tokuhisa | 3 | 0 | 0.34 |
Yasumasa Joti | 4 | 0 | 0.34 |
Kensuke Okada | 5 | 0 | 0.34 |
Takashi Sugimoto | 6 | 0 | 0.34 |
Mitsuhiro Yamaga | 7 | 1 | 0.78 |
Takaki Hatsui | 8 | 0 | 0.34 |
Makina Yabashi | 9 | 0 | 1.35 |
Yuji Sugita | 10 | 0 | 0.34 |
Yutaka Ishikawa | 11 | 1449 | 188.06 |
Nobuhiro Go | 12 | 0 | 0.34 |