Is there any point in contiuing with profiling the sn_node performance?
These figures need to be taken with a huge pinch of salt anyway - its only one run on one machine.
If you feel there is any value then I would run the smaller tests say 5 times each and take an average.
filesize | sn_node 0.48.1 | sn_node 0.49.0 | sn_node 0.49.2 | sn_node 0.49.7 |
---|---|---|---|---|
5MB | 2m1.468s | 0m14.686s | 0m14.785s | 0m18.793s |
10MB | 3m43.119s | 0m35.215s | 0m36.218s | 0m45.235s |
20MB | 6m59.187s | 1m29.119s | 1m39.737s | 1m51.249s |
50MB | 17m25.503s | 5m40.328s | 5m48.750s | 6m39.017s |
100MB | 41m34.945s | 15m53.138s | 16m27.564s | 21m20.357s |
200MB | 110m25.682s | 61m19.929s | 73m21.797s | 73m17.555s |
512MB | 368m8.248s | aborted | 388m26.748s | aborted |