Informal benchmarking

[dbwiddis]

The "is it worth it" thread. Informal benchmarks:

macOS process implementation: 10000 reps of getProcesses() with ~500 processes.

• OSHI avg: 38.26 ms
• OSHI-FFM avg: 7.23 ms (5.3 x faster)
• OSHI parallel avg: 10.23 ms
• OSHI-FFM parallel avg: 2.35 ms (4.4 x faster)

[adrian-kong]

wait that's pretty fast! would it be useful to create a triggerable benchmark workflow that pulls oshi as well as oshi ffm and runs on both just for logging?

[dbwiddis]

Note the “informal” in the title. :)

Not sure how useful a formal benchmark would be. Too many variables. Feel free to try if you want.