dualvtable/dcgmproftester repository overview

Usage: docker run --rm --gpus all --cap-add=SYS_ADMIN -t dualvtable/dcgmproftester:2.0.10-cuda11.0-ubuntu18.04 --no-dcgm-validation -t 1004 -d 5

where -t represents the type of workload to be generated (fieldID in DCGM parlance) and -d represents the duration (in seconds) that the load should be generated on the GPU.