0.20.27

Backends

Slurm

This release adds Slurm as a new backend, allowing dstack to orchestrate container-based runs across your existing Slurm clusters. A single slurm backend can manage one or many clusters — dstack connects to each cluster's login node over SSH and submits runs as Slurm jobs, with every cluster becoming its own dstack region.

projects:
- name: main
  backends:
    - type: slurm
      clusters:
        - name: gpu-cluster-a
          hostname: login.example.com
          user: admin
          private_key:
            path: ~/.ssh/id_rsa
          gpu_partitions:
            - gpu: H100
              partitions: [gpu]

Refer to the Backends docs for prerequisites, partition configuration, and known limitations.

Runs

CUDA 13.0

dstack base Docker images have CUDA runtime updated from version 12.8 to 13.0.

Note: CUDA 13 dropped support for Nvidia Volta (V100) and older architectures. If your workload depends on dstack-shipped CUDA 12.8 runtime, install the CUDA runtime manually or use a custom Docker image.

Breaking changes

• Dropped P100 provisioning from GCP and OCI.

Deprecations

• dstack gateway update: the --domain argument is deprecated in favor of gateway configuration in-place update.
• /api/project/{project_name}/gateways/create and /api/project/{project_name}/gateways/set_wildcard_domain API methods are deprecated in favor of plan-based API: /api/project/{project_name}/gateways/get_plan and /api/project/{project_name}/gateways/apply

What's Changed

• Rework the landing page layout (React/Cloudscape) by @peterschmidt85 in #3987
• [Docs] Document gRPC mode for PD disaggregation by @Bihan in #3968
• Mixing CoreModel variants in diff_models() by @jvstme in #3993
• [chore]: Fix diff_models() unit test by @jvstme in #3994
• Update configuration handling in gateway tests by @jvstme in #3995
• Add gateway replica statuses and pipeline by @jvstme in #3990
• Add in-place update for gateway domain by @jvstme in #3997
• Fix TestGetRunsTable::test_simple_run by @jvstme in #3999
• Fix/ebs device resolve retry by @james-boydell in #4002
• Update Cuda runtime from 12.8 to 13.0 in dstack base Docker images by @r4victor in #4007
• Fix CloudRift SSH port mapping by @peterschmidt85 in #4005
• Kubernetes: set NVIDIA_VISIBLE_DEVICES=void for CPU-only pods by @un-def in #4011
• Retry NVIDIA containers without display capability by @peterschmidt85 in #4006
• Add Slurm backend by @un-def in #4009
• [Landing] Minor update to landing by @peterschmidt85 in #4000
• Use yaml.safe_load everywhere by @un-def in #4015
• Slurm: fix memory request by @un-def in #4016
• Drop cudo backend by @r4victor in #4018

Full Changelog: 0.20.26...0.20.27