Skip to main content
Microsoft
separator
https://catalogartifact.azureedge.net/publicartifacts/paratools-inc.pt-pro-4-e4s-msa-oddc-amd64-d119f767-5de1-4a15-a50d-c2ae63a33920/image2_paratoolspro250250.png

ParaTools Pro for E4S™: AI/ML & HPC Tools on Heidi (x86_64)

by ParaTools, Inc.

Free Trial

ParaTools Pro for E4S™ is a turnkey HPC Cluster solution with tuned MPI, batch job management, and a host of AI/ML and science & engineering tools.

ParaTools Pro for E4S™[1] - the Extreme-scale Scientific Software Stack[2] hardened for commercial clouds and supported by ParaTools, Inc. - provides a platform for developing and deploying HPC and AI/ML applications on clusters provisioned by Adaptive Computing's Heidi. It features a browser-based (HTML5/WebRTC) remote desktop on the login node and compute nodes interconnected by the low-latency, high-bandwidth MVAPICH MPI (MVAPICH4-Plus) tuned for Microsoft Azure's InfiniBand-enabled VMs. ParaTools Pro for E4S™ ships a suite of over 100 HPC tools from E4S 26.06, built on Ubuntu 26.04 LTS with the Spack[3] package manager: ready-to-use HPC applications (such as OpenFOAM, LAMMPS, GROMACS, Quantum ESPRESSO, CP2K, WRF, Nek5000, and Xyce), numerical libraries (such as PETSc, Trilinos, hypre, SuperLU-Dist, SUNDIALS, Kokkos, and RAJA), and performance tools (such as TAU, HPCToolkit, and PAPI). AI/ML tools run in a unified CUDA-enabled Python environment: PyTorch, TensorFlow, Keras, JAX, vLLM, SGLang, Hugging Face Transformers, NVIDIA NeMo™ and BioNeMo frameworks, Megatron, Ray, XGBoost, scikit-learn, OpenCV, matplotlib, and JupyterLab with Jupyter notebooks, plus Ollama for local LLM serving, with one-command installers for the NVIDIA NemoClaw and OpenClaw agentic AI frameworks. Multi-GPU and multi-node communication is accelerated by NVIDIA NCCL, NVSHMEM, and NIXL. Julia (with CUDA and MPI support), the VSCodium IDE, ParaView, VisIt, Docker, Podman, Singularity, and Kubernetes (K3s) round out the platform. New packages can be installed easily with Spack and pip on the cluster login and compute nodes. Use it to develop the next generation of HPC engineering and science applications and generative AI applications. E4S™ has built a unified computing environment for deployment of open-source projects. E4S™ was originally developed to provide a common software environment for the exascale leadership computing systems currently being deployed at DOE National Laboratories across the U.S. Support for ParaTools Pro for E4S™ is available through ParaTools, Inc. This product has additional charges associated with it for optional product support and updates. Users wanting to stand up an HPC cluster with a complete software stack for AI/ML and/or HPC engineering and science applications will benefit from this offer. Building and configuring software for a consistent, robust, and performant environment is surprisingly difficult. This product addresses the need for a complete software stack for HPC and AI/ML applications, including a remote desktop environment, batch job management, and a host of AI/ML and science & engineering tools. This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research (ASCR), under SBIR Award Number DE-SC0022502 ("E4S: Extreme-Scale Scientific Software Stack for Commercial Clouds"). Note: This product contains repackaged and tuned open source software, e.g., E4S™, Spack, and AI/ML tools like NVIDIA NeMo™, PyTorch, JAX, Keras, etc. The full list of E4S applications installed via Spack is as follows: adios, adios2, alquimia, aml, amrex, arborx, argobots, ascent, axom, boost, bricks, butterflypack, cabana, caliper, chai, charliecloud, conduit, cp2k, cusz, darshan-runtime, darshan-util, datatransferkit, dyninst, e4s-alc, e4s-cl, faodel, flit, flux-core, fpm, gasnet, ginkgo, globalarrays, gmp, gotcha, gptune, gromacs, h5bench, hdf5, hdf5-vol-async, hdf5-vol-cache, hdf5-vol-log, heffte, hpctoolkit, hpx, hypre, kokkos, kokkos-kernels, laghos, lammps, lbann, legion, libcatalyst, libceed, libnrm, libpressio, libquo, libunwind, loki, magma, mercury, metall, mfem, mgard, mpark-variant, mpifileutils, nccmp, nco, nek5000, nekbone, netcdf-fortran, netlib-scalapack, nrm, omega-h, openfoam, openmpi, openpmd-api, papi, papyrus, parallel-netcdf, parsec, pdt, petsc, phist, plasma, plumed, precice, pruners-ninja, pumi, py-cinemasci, py-h5py, py-jupyterhub, py-libensemble, py-petsc4py, qthreads, quantum-espresso, raja, rempi, scr, slate, slepc, strumpack, sundials, superlu, superlu-dist, swig, sz, sz3, tasmanian, tau, trilinos, turbine, umap, umpire, upcxx, variorum, veloc, vtk-m, wannier90, warpx, wps, wrf, xyce, zfp [1]: https://paratoolspro.com/ [2]: https://e4s.io [3]: https://spack.io
English (United States)
Your Privacy Choices Opt-Out Icon Your Privacy Choices
Consumer Health Privacy Sitemap Contact Us Privacy & Cookies Terms of Use Trademarks About our ads Manage cookies