Abstract | ||
---|---|---|
Conventional auto-tuning numerical software has the drawbacks of (1) fixed sampling points for the performance estimation; (2) inadequate adaptation to heterogeneous environments. To solve these drawbacks, we developed ABCLib_DRSSED, which is a parallel eigensolver with an auto-tuning facility. ABCLib_DRSSED has (1) functions based on the sampling points which are constructed with an end-user interface; (2) a load-balancer for the data to be distributed; (3) a new auto-tuning optimization timing called Before Execute-time Optimization (BEO). In our performance evaluation of the BEO, we obtained speedup factors from 10% to 90%, and 340% in the case of a failed estimation. In the evaluation of the load-balancer, the performance was 220% improved. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1016/j.parco.2005.10.002 | Parallel Computing |
Keywords | Field | DocType |
new auto-tuning optimization timing,failed estimation,sampling point,auto-tuning facility,fixed sampling point,performance estimation,fiber,execute-time optimization,end-user interface,parallel eigensolver,abclib,conventional auto-tuning numerical software,performance evaluation,load-balancer,user interface,load balance,load balancer,numerical software | Load balancing (computing),Computer science,Parallel computing,Performance estimation,Sampling (statistics),Numerical analysis,Auto tuning,Speedup | Journal |
Volume | Issue | ISSN |
32 | 3 | Parallel Computing |
Citations | PageRank | References |
13 | 0.91 | 11 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Takahiro Katagiri | 1 | 121 | 17.01 |
Kenji Kise | 2 | 149 | 26.53 |
Hiroki Honda | 3 | 84 | 8.72 |
Toshitsugu Yuba | 4 | 265 | 37.72 |