Open arthurpaulino opened 11 months ago
This is stated in far too definite a form. It might or might not be helpful in the context of an optimized pipeline on a real system.
@porcuquine please feel free to edit 🙏🏼
I think my comment covers the nuance that was missing. In terms of substance:
Moreover, this would put witness generation in competition with MSM for GPU usage. Currently, our best-performing proof generation shows negligible CPU usage, so the most promising approach to optimal throughput is likely better CPU utilization to minimize witness-generation time.
The witness generation for the slots involving Poseidon hashing could be performed on GPU, which would probably provide further improvements on top of the parallelization implemented in #766.