| layout | single | ||
|---|---|---|---|
| title | E4S Container Download | ||
| redirect_from |
|
||
| permalink | /container-download/ | ||
| classes | wide | ||
| sidebar |
|
{% include e4s-page-actions.html %}
The current E4S container offerings include Docker and Singularity images capable of running on X86_64, PPC64LE, and AARCH64 architectures. Our full E4S Release images are based on Ubuntu 24.04 and Rocky 9.6 (x86_64, aarch64, ppc64le). In addition to offering a full E4S image containing a comprehensive selection of E4S software released on a bi-annual cycle, we also offer a set of minimal base images suitable for use in Continuous Integration (CI) pipelines where Spack is used to build packages.
Docker images are available on the E4S Docker Hub. Please see the E4S 25.11 Release Notes.
- Docker Downloads - Minimal
- Docker Downloads - CPU only - Ubuntu
- Docker Downloads - CPU only - Rocky
- Docker Downloads - CUDA - Ubuntu
- Docker Downloads - CUDA - Rocky
- Docker Downloads - ROCm
- Docker Downloads - OneAPI
- Singularity x86_64 Download - Minimal
- Singularity x86_64 Download - CPU only - Ubuntu
- Singularity x86_64 Download - CPU only - Rocky
- Singularity x86_64 Download - CUDA 80 - Ubuntu
- Singularity x86_64 Download - CUDA 90 - Ubuntu
- Singularity x86_64 Download - CUDA 90 - Rocky
- Singularity x86_64 Download - CUDA 120 - Ubuntu
- Singularity x86_64 Download - CUDA 120 - Rocky
- Singularity x86_64 Download - ROCm gfx942 - Ubuntu
- Singularity x86_64 Download - ROCm gfx90a - Ubuntu
- Singularity x86_64 Download - ROCm gfx908 - Ubuntu
- Singularity x86_64 Download - OneAPI - Ubuntu
- Singularity ppc64le Download - CUDA 70 - Ubuntu
- Singularity aarch64 Download - CPU only - Ubuntu
- Singularity aarch64 Download - CUDA 80 - Ubuntu
- Singularity aarch64 Download - CUDA 90 - Ubuntu
- Singularity aarch64 Download - CUDA 90 - Rocky
- Singularity aarch64 Download - CUDA 120 - Ubuntu
- Singularity aarch64 Download - CUDA 120 - Rocky
- OVA Download
Container images contain binary versions of the Full Release packages listed above. Full-featured GPU-enabled container images are available from Dockerhub:
# docker pull ecpe4s/e4s-cuda:25.11-cuda120
# docker pull ecpe4s/e4s-cuda:25.11-cuda90
# docker pull ecpe4s/e4s-cuda:25.11-cuda80
# docker pull ecpe4s/e4s-cuda:25.11-cuda70
# docker pull ecpe4s/e4s-cuda-rocky9.6:25.11-cuda120
# docker pull ecpe4s/e4s-cuda-rocky9.6:25.11-cuda90
# docker pull ecpe4s/e4s-rocm:25.11-rocm942
# docker pull ecpe4s/e4s-rocm:25.11-rocm90a
# docker pull ecpe4s/e4s-rocm:25.11-rocm908
# docker pull ecpe4s/e4s-oneapi:25.11
# docker pull ecpe4s/e4s-cpu:25.11
# docker pull ecpe4s/e4s-cpu-rocky9.6:25.11These images contain a full Spack-based deployment of E4S, including GPU-enabled packages for NVIDIA, AMD, or Intel GPUs.
These images also contain NVIDIA NeMo, NVIDIA BioNeMo, VLLM, PyTorch, TensorFlow, and TAU.
- ecpe4s/e4s-rocm:25.11 [docker]
- e4s-rocm942-x86_64-25.11.sif [singularity]
- e4s-rocm90a-x86_64-25.11.sif [singularity]
- e4s-rocm908-x86_64-25.11.sif [singularity]
- ecpe4s/e4s-cuda:25.11 [docker]
- e4s-cuda80-x86_64-25.11.sif [singularity]
- e4s-cuda90-x86_64-25.11.sif [singularity]
- e4s-cuda120-x86_64-25.11.sif [singularity]
- e4s-cuda90-rocky9.6-x86_64-25.11.sif [singularity]
- e4s-cuda120-rocky9.6-x86_64-25.11.sif [singularity]
- e4s-cuda70-ppc64le-25.11.sif [singularity]
- e4s-cuda80-aarch64-25.11.sif [singularity]
- e4s-cuda90-aarch64-25.11.sif [singularity]
- e4s-cuda120-aarch64-25.11.sif [singularity]
- e4s-cuda90-rocky9.6-aarch64-25.11.sif [singularity]
- e4s-cuda120-rocky9.6-aarch64-25.11.sif [singularity]
- ecpe4s/e4s-oneapi:25.11 [docker]
- e4s-oneapi-x86_64-25.11.sif [singularity]
- ecpe4s/e4s-cpu:25.11 [docker]
- e4s-cpu-x86_64-25.11.sif [singularity]
- e4s-cpu-rocky9.6-x86_64-25.11.sif [singularity]
- e4s-cpu-aarch64-25.11.sif [singularity]
- e4s-cpu-rocky9.6-aarch64-25.11.sif [singularity]
These images come with MPICH, CMake, and the relevant GPU SDK -- either AMD ROCm, NVIDIA CUDA Toolkit and NVHPC, or Intel OneAPI.
- ecpe4s/e4s-base-rocm:25.11 [docker]
- e4s-base-rocm-25.11.sif [singularity]
- ecpe4s/e4s-base-cuda:25.11 [docker]
- e4s-base-cuda-x86_64-25.11.sif [singularity]
- e4s-base-cuda-aarch64-25.11.sif [singularity]
- e4s-base-cuda-ppc64le-25.11.sif [singularity]
- ecpe4s/e4s-base-oneapi:25.11 [docker]
- e4s-base-oneapi-25.11.sif [singularity]
This image contains a minimal setup for using Spack 0.22.0 w/ GNU compilers
- ecpe4s/ubuntu20.04 [docker]
- ecpe4s-ubuntu20.04-x86_64-24.02.sif [singularity]
- ecpe4s-ubuntu20.04-ppc64le-24.02.sif [singularity]
- ecpe4s-ubuntu20.04-aarch64-24.02.sif [singularity]
This multi-architecture image contains E4S products compiled with DOE LLVM 16 and Flang using Spack
- ecpe4s/e4s-doe-llvm:23.05 [docker]
- e4s-doe-llvm-x86_64-23.05.sif [singularity]
- e4s-doe-llvm-aarch64-23.05.sif [singularity]
- e4s-doe-llvm-ppc64le-23.05.sif [singularity]
- ecpe4s/sollve-rocm6.3.0 [docker]
- ecpe4s/sollve-cuda12.6.3-arm64 [docker]
- ecpe4s/sollve-cuda12.6.3-amd64 [docker]
- ecpe4s/trilinos-sycl [docker]
- ecpe4s/trilinos-ci-rocm6.2.1 [docker]
- ecpe4s/trilinos-ci-cuda12.2.2 [docker]
- ecpe4s/trilinos-ci-arm64-cuda12 [docker]
- ecpe4s/exawind-snapshot [docker]
- ecpe4s/exago-cuda80 [docker]
- ecpe4s/exago-rocm90a [docker]
- ecpe4s/ubuntu22.04-runner-x86_64 [docker] [github]
- ecpe4s/ubuntu20.04-runner-x86_64 [docker] [github]
- ecpe4s/ubuntu18.04-runner-x86_64 [docker] [github]
- ecpe4s/rhel8-runner-x86_64 [docker] [github]
- ecpe4s/rhel7-runner-x86_64 [docker] [github]
- ecpe4s/ubuntu22.04-runner-ppc64le [docker] [github]
- ecpe4s/ubuntu20.04-runner-ppc64le [docker] [github]
- ecpe4s/ubuntu18.04-runner-ppc64le [docker] [github]
- ecpe4s/rhel8-runner-ppc64le [docker] [github]
- ecpe4s/rhel7-runner-ppc64le [docker] [github]
- ecpe4s/ubuntu22.04-runner-aarch64 [docker] [github]
- ecpe4s/ubuntu20.04-runner-aarch64 [docker] [github]
- ecpe4s/rhel8-runner-aarch64 [docker] [github]
The E4S 25.11 release is also available on AWS as an EC2 AMI with ID ami-0e752117cfa13cb9b in the US-West-2 (Oregon) region.