Difference between revisions of "Performance"
From Gw-qcd-wiki
(Created page with ''''Carver''' Tester: Ben Gamari Test date: 14 Jul 2010 Commit: e3e4ffafd158abd004c483694a27f4f6bc7d2185 Hardware: CUDA version 3.0 {| !Kernel !Configuration !Bandwidth !FLOPs…') |
|||
| Line 7: | Line 7: | ||
{| | {| | ||
| − | !Kernel !Configuration !Bandwidth !FLOPs | + | !Kernel |
| + | !Configuration | ||
| + | !Bandwidth | ||
| + | !FLOPs | ||
|- | |- | ||
|rowspan="2"|Dslash_cuda | |rowspan="2"|Dslash_cuda | ||
Revision as of 13:33, 14 July 2010
Carver Tester: Ben Gamari Test date: 14 Jul 2010 Commit: e3e4ffafd158abd004c483694a27f4f6bc7d2185 Hardware: CUDA version 3.0
| Kernel | Configuration | Bandwidth | FLOPs |
|---|---|---|---|
| Dslash_cuda | 73 GB/s |32 GFLOP/s | ||
| 74 GB/s |34 GFLOP/s | |||
| Dslash_multi_gpu (double) | 79 GB/s |35 GFLOP/s | ||
| 145 GB/s |64 GFLOP/s | |||
| 256 GB/s |114 GFLOP/s | |||
| Dslash_multi_gpu (double) | 79 GB/s |76 GFLOP/s | ||
| 156 GB/s |140 GFLOP/s | |||
| 283 GB/s |252 GFLOP/s | |||
| 82 GB/s |3.4 GFLOP/s | |||
| 88 GB/s |N/A | |||
| 84 GB/s |N/A |
