Acquiring E4S Containers
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 22.04 (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 24.11 Release Notes.
Container Releases
Singularity x86_64 Download - CUDA 80
Singularity x86_64 Download - CUDA 90
Singularity ppc64le Download - CUDA 70Singularity aarch64 Download - CUDA 75
Singularity aarch64 Download - CUDA 80
Singularity aarch64 Download - CUDA 90
Singularity x86_64 Download - ROCm gfx942
Singularity x86_64 Download - ROCm gfx90a
Singularity x86_64 Download - ROCm gfx908 Singularity x86_64 Download - OneAPIFrom source with Spack
Spack contains packages for all of the products listed in the E4S 24.11 Full Release category (see above Release Notes). General instructions for building software with Spack can be found at the Spack website. Questions concerning building those packages are deferred to the associated package development team.
Note on Container Images
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:24.11
# docker pull ecpe4s/e4s-rocm:24.11
# docker pull ecpe4s/e4s-oneapi:24.11
E4S Full GPU Images
These images contain a full Spack-based deployment of E4S, including GPU-enabled packages for NVIDIA, AMD, or Intel GPUs.
These images also contain TensorFlow, PyTorch, and TAU.
AMD ROCm (x86_64)
NVIDIA CUDA (X86_64, PPC64LE, AARCH64)
Intel OneAPI (x86_64)
- ecpe4s/e4s-oneapi:24.11
- e4s-oneapi-x86_64-24.11.sif
mirror 1
GPU Base Images
These images come with MPICH, CMake, and the relevant GPU SDK -- either AMD ROCm, NVIDIA CUDA Toolkit and NVHPC, or Intel OneAPI.
AMD ROCM (X86_64)
- ecpe4s/e4s-base-rocm:24.11
- e4s-base-rocm-24.11.sif mirror 1
NVIDIA Multi-Arch (X86_64, PPC64LE, AARCH64)
Intel OneAPI (X86_64)
- ecpe4s/e4s-base-oneapi:24.11
- e4s-base-oneapi-24.11.sif mirror 1
DOE LLVM E4S Image
This multi-architecture image contains E4S products compiled with DOE LLVM 16 and Flang using Spack
Multi-Arch (X86_64, PPC64LE, AARCH64)
Continuous Integration Images
These are barebones operating system images which contain only essential build tools and python packages needed by Spack.
These images are intended to be used in continuous integration workflows where Spack is first cloned and then used to build and test software.
X86_64
- ecpe4s/ubuntu22.04-runner-x86_64
- ecpe4s/ubuntu20.04-runner-x86_64
- ecpe4s/ubuntu18.04-runner-x86_64
- ecpe4s/rhel8-runner-x86_64
- ecpe4s/rhel7-runner-x86_64
PPC64LE
- ecpe4s/ubuntu22.04-runner-ppc64le
- ecpe4s/ubuntu20.04-runner-ppc64le
- ecpe4s/ubuntu18.04-runner-ppc64le
- ecpe4s/rhel8-runner-ppc64le
- ecpe4s/rhel7-runner-ppc64le
AARCH64
- ecpe4s/ubuntu22.04-runner-aarch64
- ecpe4s/ubuntu20.04-runner-aarch64
- ecpe4s/rhel8-runner-aarch64
Custom Images
- ecpe4s/waggle-ml
- ecpe4s/exawind-snapshot
- ecpe4s/superlu_sc
AWS EC2 Image
The E4S 24.11 release is also available on AWS as an EC2 AMI with ID ami-0e752117cfa13cb9b in the US-West-2 (Oregon) region.