Optimized 1024 * 1024 matrix multiplication in pure C from naive to tiling to SIMD avx register on my i3 2nd gen CPU and benchmark it against OpenBlas and here's the result
Feel free to explore, modify, benchmark, or improve it.
7
35
10
27.3K
Snapshot History (25)
| Date | ||||
|---|---|---|---|---|
| Jul 1 · 16:00 | 7 | +1 35 | 10 | +282 27.3K |
| Jul 1 · 15:00 | 7 | 34 | 10 | +209 27.0K |
| Jul 1 · 14:00 | 7 | 34 | 10 | +207 26.8K |
| Jul 1 · 13:00 | 7 | +1 34 | +2 10 | +473 26.6K |
| Jul 1 · 11:00 | 7 | 33 | 8 | +221 26.1K |
| Jul 1 · 10:00 | 7 | 33 | 8 | +224 25.9K |
| Jul 1 · 09:00 | 7 | +1 33 | +1 8 | +418 25.6K |
| Jul 1 · 07:00 | 7 | 32 | 7 | +276 25.2K |
| Jul 1 · 06:00 | 7 | 32 | 7 | +266 24.9K |
| Jul 1 · 05:00 | 7 | 32 | 7 | +424 24.7K |
| Jul 1 · 04:00 | 7 | 32 | 7 | +413 24.3K |
| Jul 1 · 03:00 | 7 | 32 | 7 | +396 23.8K |
| Jul 1 · 02:00 | 7 | 32 | 7 | +526 23.5K |
| Jul 1 · 01:00 | 7 | 32 | 7 | +1.0K 22.9K |
| Jun 30 · 23:00 | 7 | 32 | 7 | +642 21.9K |
| Jun 30 · 22:00 | 7 | 32 | 7 | +737 21.3K |
| Jun 30 · 21:00 | 7 | 32 | 7 | +770 20.5K |
| Jun 30 · 20:00 | 7 | 32 | 7 | +762 19.8K |
| Jun 30 · 19:00 | 7 | 32 | 7 | +907 19.0K |
| Jun 30 · 18:00 | 7 | 32 | 7 | +929 18.1K |
| Jun 30 · 17:00 | 7 | +1 32 | 7 | +759 17.2K |
| Jun 30 · 16:00 | +1 7 | +4 31 | +1 7 | +3.2K 16.4K |
| Jun 30 · 12:00 | +1 6 | +1 27 | +1 6 | +708 13.2K |
| Jun 30 · 11:00 | 5 | 26 | 5 | +665 12.5K |
| Jun 30 · 10:00 | +5 5 | +26 26 | +5 5 | +11.8K 11.8K |
| Jun 29 · 15:47 | 0 | 0 | 0 | 0 |