Table 3 Execution time (in seconds) of parallel ISS based tiled code for some tile sizes, N = 5000, 32 threads used
From: Tuning iteration space slicing based tiled multi-core code implementing Nussinov’s RNA folding
No. | b1 | b2 | b3 | Time |
|---|---|---|---|---|
1 | 1 | 96 | 8 | 7.8751 |
2 | 1 | 150 | 12 | 8.0246 |
3 | 1 | 96 | 12 | 8.1903 |
4 | 1 | 128 | 12 | 8.1952 |
5 | 1 | 128 | 16 | 8.2199 |
6 | 1 | 128 | 6 | 8.2816 |
7 | 1 | 150 | 16 | 8.2831 |
8 | 1 | 50 | 8 | 8.3449 |
9 | 1 | 128 | 8 | 8.3841 |
10 | 1 | 96 | 6 | 8.4597 |
... | ... | ... | ... | ... |
143 | 2 | 6 | 300 | 10.4351 |
... | ... | ... | ... | ... |
Execution time of the original code: | 334.3200 | |||
7993 | 2 | 1 | 2 | 335.3765 |
7994 | 1 | 2 | 2 | 411.5945 |
... | ... | ... | ... | ... |
8000 | 1 | 2 | 1 | 889.6510 |