diff --git a/source/guides/azure/infiniband.md b/source/guides/azure/infiniband.md index daca2391..a65840e3 100644 --- a/source/guides/azure/infiniband.md +++ b/source/guides/azure/infiniband.md @@ -252,7 +252,7 @@ Accept the default and allow conda init to run. Then start a new shell. Create a conda environment (see [UCX-Py](https://ucx-py.readthedocs.io/en/latest/install.html) docs) ```shell -mamba create -n ucxpy {{ rapids_conda_channels }} {{ rapids_conda_packages }} ipython ucx-proc=*=gpu ucx ucx-py dask distributed numpy cupy pytest pynvml -y +mamba create -n ucxpy {{ rapids_conda_channels }} {{ rapids_conda_packages }} ipython ucx-proc=*=gpu ucx ucx-py dask distributed numpy cupy pytest nvidia-ml-py -y mamba activate ucxpy ``` diff --git a/source/guides/mig.md b/source/guides/mig.md index f2e3ad6d..cb01d5ae 100644 --- a/source/guides/mig.md +++ b/source/guides/mig.md @@ -30,7 +30,7 @@ GPU 0: NVIDIA A100-PCIE-40GB (UUID: GPU-84fd49f2-48ad-50e8-9f2e-3bf0dfd47ccb) In the example case above the system has one NVIDIA A100 with 3 x 10GB MIG instances. In the next sections we will see how to use the instance names to startup a Dask cluster composed of MIG GPUs. Please note that once a GPU is partitioned, the physical GPU (named `GPU-84fd49f2-48ad-50e8-9f2e-3bf0dfd47ccb` above) is inaccessible for CUDA compute and cannot be used as part of a Dask cluster. -Alternatively, MIG instance names can be obtained programatically using [NVML](https://developer.nvidia.com/nvidia-management-library-nvml) or [PyNVML](https://github.com/gpuopenanalytics/pynvml). Please refer to the [NVML API](https://docs.nvidia.com/deploy/nvml-api/) to write appropriate utilities for that purpose. +Alternatively, MIG instance names can be obtained programatically using [NVML](https://developer.nvidia.com/nvidia-management-library-nvml) or [nvidia-ml-py](https://pypi.org/project/nvidia-ml-py/). Please refer to the [NVML API](https://docs.nvidia.com/deploy/nvml-api/) to write appropriate utilities for that purpose. ### LocalCUDACluster