You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Running some GPU job challenge directly on the bare machine can be unreliable given random differences in environment libs and CUDA versions. We should create a dockerized benchmark that can be queued as a job by the miner, something like gpu-burn, that is fed in challenge parameters as a file and writes output to the same folder once it's done.
The text was updated successfully, but these errors were encountered:
Running some GPU job challenge directly on the bare machine can be unreliable given random differences in environment libs and CUDA versions. We should create a dockerized benchmark that can be queued as a job by the miner, something like
gpu-burn
, that is fed in challenge parameters as a file and writes output to the same folder once it's done.The text was updated successfully, but these errors were encountered: