Comparison tests of running GapMiner on the GPU at different queue size settings. The tabular data are the more communicative of the relative differences in performance.
Column labels map directly to miner output: pps is average primes per second, tps is average tests per second, gps is average gaps per second, glst is the size of the gaplist and l/s is the time in seconds to scan the gaplist at the reported rate of gaps per second.
f64-t1-i3000000-s12000000-z5-n8-w512
| measure | pps | tps |
|---|---|---|
| mean | 2722038 | 2055509 |
| max | 3120218 | 2359843 |
| min | 2672679 | 2018498 |
| std | 59997 | 45482 |
f64-t1-i3000000-s12000000-z10-n8-w512
| measure | pps | tps |
|---|---|---|
| mean | 2725573 | 2057521 |
| max | 3178352 | 2392060 |
| min | 2674148 | 2018796 |
| std | 74038 | 55660 |
f64-t1-i3000000-s12000000-z20-n8-w512
| measure | pps | tps |
|---|---|---|
| mean | 2640409 | 1994529 |
| max | 2683620 | 2027112 |
| min | 2453759 | 1844421 |
| std | 33151 | 25871 |
f64-t1-i3000000-s12000000-z30-n8-w512
| measure | pps | tps |
|---|---|---|
| mean | 2687596 | 2028362 |
| max | 2741437 | 2069912 |
| min | 2587098 | 1947674 |
| std | 32710 | 25326 |
The charts included below are for completeness. The figures given are the means and they differ from the pandas-produced means because, in order to support the different requirements of charting, the first 1/5th of the data is ignored in order to aid visual comparison. The chartline uses the full dataset.
Replication: