EPIC: Port Rapids projects from pynvml to cuda.core.system

[mdboom]

This is an EPIC to track porting usage of pynvml to cuda.core.system across the RAPIDS family of projects. In order of importance / level of NVML usage.

• dask_cuda

  • cuda.core.system: Add MIG-related APIs #1916
  • Migrate from pynvml to cuda.core.system rapidsai/dask-cuda#1645

• cudf

  • cuda.core.system: Add MIG-related APIs #1916
  • cuda.core.system: Add ProcessInfo APIs #1917
  • https://github.com/rapidsai/cudf/compare/main...mdboom:cudf:pynvml-to-cuda.core.system?expand=1

• jupyterlab-nvdashboard

  • cuda.core.system: Add basic Nvlink and Utilization support #1918
  • https://github.com/rapidsai/jupyterlab-nvdashboard/compare/branch-0.14...mdboom:pynvml-to-cuda.core.system?expand=1

• rapids-cli

  • cuda.core.system: Add basic Nvlink and Utilization support #1918
  • https://github.com/rapidsai/rapids-cli/compare/main...mdboom:rapids-cli:pynvml-to-cuda.core.system?expand=1

• deployment

  • Mention cuda.core.system as an option for getting NVML information. rapidsai/deployment#677

• ucxx

  • cuda.core.system: Add MIG-related APIs #1916
  • https://github.com/rapidsai/ucxx/compare/main...mdboom:ucxx:pynvml-to-cuda.core.system?expand=1

• auto-nvidia-accelerator (This should have been archived)

  • https://github.com/rapidsai/auto-nvidia-accelerator/pull/182

• rapidsmpf

  • Migrate from pynvml to cuda.core.system rapidsai/rapidsmpf#969

• cuml

  • pynvml is only used in testing. Deferring.

• cugraph

  • pynvml is only used in a developer utility. Deferring.

• kvikio

  • pynvml is used for benchmarking only. Deferring

• rapids-examples

  • pynvml is used in a single example. Deferring.

• tpm-tools

  • Seems like examples? Not entirely sure. Deferring.