Release 25.02 Checklist

[jacobtomlinson]

Release checklist

For the upcoming release we need to verify our documentation. This is a best efforts activity
so please refer to the checklist from the previous release and focus on pages that were not
verified last time.

Verify pages

• Look at the nightly build of each page listed below
• Check page renders correctly
• Check for spelling/grammar problems
• Check that the instructions work as expected
• Ensure legacy pages with out of date instructions have a content warning
• If page needs updating convert the task to an issue and open a PR that closes the issue

Index/Non-technical

• https://docs.rapids.ai/deployment/nightly/
• https://docs.rapids.ai/deployment/nightly/cloud
• https://docs.rapids.ai/deployment/nightly/cloud/aws
• https://docs.rapids.ai/deployment/nightly/cloud/azure
• https://docs.rapids.ai/deployment/nightly/cloud/gcp
• https://docs.rapids.ai/deployment/nightly/cloud/ibm
• https://docs.rapids.ai/deployment/nightly/cloud/nvidia
• https://docs.rapids.ai/deployment/nightly/developer
• https://docs.rapids.ai/deployment/nightly/developer/ci
• https://docs.rapids.ai/deployment/nightly/examples
• https://docs.rapids.ai/deployment/nightly/guides
• https://docs.rapids.ai/deployment/nightly/hpc
• https://docs.rapids.ai/deployment/nightly/local
• https://docs.rapids.ai/deployment/nightly/nims
• https://docs.rapids.ai/deployment/nightly/platforms
• https://docs.rapids.ai/deployment/nightly/tools

P0

• https://docs.rapids.ai/deployment/nightly/cloud/aws/sagemaker
  • but had to pin to RAPIDS 24.12: use 'sagemaker-python-sdk' instead of 'sagemaker' #504
• https://docs.rapids.ai/deployment/nightly/cloud/azure/azureml
• https://docs.rapids.ai/deployment/nightly/cloud/gcp/vertex-ai
• https://docs.rapids.ai/deployment/nightly/platforms/colab
• https://docs.rapids.ai/deployment/nightly/platforms/databricks

P1

• https://docs.rapids.ai/deployment/nightly/cloud/aws/ec2
• https://docs.rapids.ai/deployment/nightly/cloud/aws/eks
• https://docs.rapids.ai/deployment/nightly/cloud/azure/aks
• https://docs.rapids.ai/deployment/nightly/cloud/azure/azure-vm
• https://docs.rapids.ai/deployment/nightly/cloud/gcp/compute-engine
• https://docs.rapids.ai/deployment/nightly/cloud/gcp/gke

P2

• https://docs.rapids.ai/deployment/nightly/cloud/nvidia/brev
• https://docs.rapids.ai/deployment/nightly/cloud/aws/ec2-multi
• https://docs.rapids.ai/deployment/nightly/cloud/aws/ecs
• https://docs.rapids.ai/deployment/nightly/cloud/azure/azure-vm-multi
• https://docs.rapids.ai/deployment/nightly/cloud/gcp/dataproc
• https://docs.rapids.ai/deployment/nightly/cloud/ibm/virtual-server
• https://docs.rapids.ai/deployment/nightly/developer/ci/github-actions
• https://docs.rapids.ai/deployment/nightly/examples/rapids-1brc-single-node/notebook
• https://docs.rapids.ai/deployment/nightly/examples/rapids-autoscaling-multi-tenant-kubernetes/notebook
• https://docs.rapids.ai/deployment/nightly/examples/rapids-azureml-hpo/notebook
• https://docs.rapids.ai/deployment/nightly/examples/rapids-ec2-mnmg/notebook
• https://docs.rapids.ai/deployment/nightly/examples/rapids-optuna-hpo/notebook
• https://docs.rapids.ai/deployment/nightly/examples/rapids-sagemaker-higgs/notebook
  • could not get a big enough machine to see these pass: [AWS] SageMaker: Higgs example fails on ml.p3.2xlarge: "an illegal memory access was encountered" #521
• https://docs.rapids.ai/deployment/nightly/examples/rapids-sagemaker-hpo/notebook
  • could not get a big enough machine to see these pass: [AWS] SageMaker: Hyperparameter Optimization example fails on ml.p3.2xlarge: "std::bad_alloc: out_of_memory" #523
• https://docs.rapids.ai/deployment/nightly/examples/time-series-forecasting-with-hpo/notebook
• https://docs.rapids.ai/deployment/nightly/examples/xgboost-azure-mnmg-daskcloudprovider/notebook
• https://docs.rapids.ai/deployment/nightly/examples/xgboost-dask-databricks/notebook
• https://docs.rapids.ai/deployment/nightly/examples/xgboost-gpu-hpo-job-parallel-k8s/notebook
• https://docs.rapids.ai/deployment/nightly/examples/xgboost-gpu-hpo-mnmg-parallel-k8s/notebook
• https://docs.rapids.ai/deployment/nightly/examples/xgboost-randomforest-gpu-hpo-dask/notebook
• https://docs.rapids.ai/deployment/nightly/examples/xgboost-rf-gpu-cpu-benchmark/notebook
• https://docs.rapids.ai/deployment/nightly/guides/azure/infiniband
• https://docs.rapids.ai/deployment/nightly/guides/caching-docker-images
• https://docs.rapids.ai/deployment/nightly/guides/colocate-workers
• https://docs.rapids.ai/deployment/nightly/guides/l4-gcp
• https://docs.rapids.ai/deployment/nightly/guides/mig
• https://docs.rapids.ai/deployment/nightly/guides/scheduler-gpu-optimization
• https://docs.rapids.ai/deployment/nightly/guides/scheduler-gpu-requirements
• https://docs.rapids.ai/deployment/nightly/platforms/coiled
• https://docs.rapids.ai/deployment/nightly/platforms/kserve
• https://docs.rapids.ai/deployment/nightly/platforms/kubeflow
• https://docs.rapids.ai/deployment/nightly/platforms/kubernetes
• https://docs.rapids.ai/deployment/nightly/platforms/nvidia-ai-workbench
• https://docs.rapids.ai/deployment/nightly/platforms/snowflake
• https://docs.rapids.ai/deployment/nightly/tools/dask-cuda
• https://docs.rapids.ai/deployment/nightly/tools/kubernetes/dask-helm-chart
• https://docs.rapids.ai/deployment/nightly/tools/kubernetes/dask-operator
• https://docs.rapids.ai/deployment/nightly/tools/rapids-docker

Issue text generated by scripts/gen_release_checklist_issue.py.

[ncclementi]

Update: note sure what happen, got access, though Azure ML instructions are broken see #515

Just a heads up: I tried to finish testing Azure ML but I'm having issues with access that I didn't have yesterday. I will give this another try on Monday.

[jameslamb]

I tested the SageMaker examples tonight.

Good News

RAPIDS 24.08, 24.10, and 25.02 conda packages no longer have any dependency conflicts with sagemaker-python-sdk 🎉

#443 (comment)

Bad News

RAPIDS 25.02 can't be installed on SageMaker Notebook Instances. The latest OS they support is Amazon Linux 2, which has GLIBC 2.26... RAPIDS 25.02 has a minimum of GLIBC 2.28.

• [AWS] SageMaker: RAPIDS 25.02 conda packages cannot be installed on Notebook Instances (requires GLIBC 2.28+) #520

I could not get SageMaker training examples working.... ran out of memory on ml.p3.2xlarge, and I couldn't provision anything larger like ml.p3.8xlarge:

• [AWS] SageMaker: Hyperparameter Optimization example fails on ml.p3.2xlarge: "std::bad_alloc: out_of_memory" #523
• [AWS] SageMaker: Higgs example fails on ml.p3.2xlarge: "an illegal memory access was encountered" #521