Test Drive NVIDIA H100 Tensor Core GPUs with Microway + Supermicro


Experience unprecedented performance, scalability and security for every data center

In partnership with NVIDIA + Supermicro, we’re offering an opportunity to test the latest NVIDIA H100 Tensor Core GPUs remotely to qualified customers.

The NVIDIA H100 Tensor Core GPU enables an order-of-magnitude leap for large-scale AI and HPC. With NVIDIA AI Enterprise for streamlined AI development and deployment, NVIDIA H100 accelerates everything from exascale scale workloads utilizing the dedicated Transformer Engine for trillion parameter language models, down to right-sized Multi-Instance GPU (MIG) partitions.

Test drive systems with NVIDIA H100 GPUs to experience an end-to-end AI and HPC data center platform.

Fast time-to-value

Test on a turnkey AI solution with NVIDIA AI Enterprise, NVIDIA H100, and NVIDIA-Certified Systems. Receive your login to a testing system rapidly.

Boost performance and efficiency

Boost your AI training performance, especially on large models, with NVIDIA H100’s Transformer Engine and achieve the highest efficiency and security with 2nd gen MIG and confidential computing.

Scale with ease with the latest GPUs

Service your entire AI pipeline on one infrastructure— with the increased scalability of the latest GPUs and superior AI software

Experience Enterprise-Ready AI

Utilize guaranteed support for streamlined Enterprise AI— with assurance your workloads run on not just this but any NVIDIA certified enterprise datacenter platform.

NVIDIA H100 Systems

Microway Navion 4U GPU Node based upon Supermicro AS-4125GS-TNRT/AS -4125GS-TNRT1:

  • Two NVIDIA H100s installed, with support for up to 8 or 10 NVIDIA H100 80GB PCI-E GPUs
  • 2x AMD EPYC 9000 Series “Genoa” CPUs with 96 Cores each
  • 768GB DDR5-4800Mhz memory
  • NVIDIA Connect-X InfiniBand HCAs and switching to additional nodes

NVIDIA AI Software

NVIDIA AI Software Configuration

Other Software

Optional Software Configuration

  • Compatibility with HPC Containers including: NAMD, VMD, GROMACS, LAMMPS, HOOMD-blue, Chroma, MILC, Julia, and more
  • Compatibility with AI Containers including: Caffe2, MXNet, PyTorch, TensorFlow
  • Rocky Linux and SLURM Scheduler

MPI & Compilers

MPI & Compiler Software

  • MVAPICH2 and OpenMPI
  • NVIDIA HPC SDK
  • GNU GCC Compiler Collection (multiple versions, as needed) Provides C, C++ and Fortran compilers.
  • AMD Optimizing C/C++ Compiler (AOCC) Provides C, C++, and Fortran compilers with optimizations for the latest AMD EPYC CPUs

Ready to get started? Sign up now!


Comments are closed.