Validate your performance with NVIDIA GPUs on Microway’s remote benchmarking cluster. Our team knows that the best way to secure your next HPC or AI deployment is to have validation of performance. We’re able to make available hardware and software as follows.
Microway offers a Linux-based benchmark cluster for customers to test GPU-enabled applications. The cluster includes:
- Microway NumberSmasher and Navion GPU Nodes with NVIDIA H100, NVIDIA A100, and L40S Tensor Core GPUs
- NVIDIA DGX A100 deep learning system (with NVSwitch/NVLink-connected NVIDIA A100 GPUs)
- CentOS, Rocky, or Ubuntu Linux (with support for Singularity images)
- Pre-configured GPU-enabled software packages
- Alternate test configurations available upon request.
Custom NumberSmasher (Xeon) and Navion (EPYC) GPU Nodes include
- Up to eight NVIDIA datacenter PCI-E or NVIDIA NVLink GPUs per node
- EDR and HDR InfiniBand HCAs and switching
GPU-Accelerated Applications Available for Testing
- TensorFlow with Keras
- PyTorch and Caffe2 deep learning frameworks
- RAPIDS for data science and analytics on GPUs
- nvBowtie (GPU-accelerated Bowtie2)
- MATLAB runtime (users may execute pre-compiled MATLAB applications)
- AMBER
- GAMESS
- GROMACS
- LAMMPS
- NAMD (with multicore, MPI and CUDA builds)
- Quantum Espresso
- RELION
- TeraChem
- VMD
- HOOMD-blue (single and double precision)
Available Libraries & SDKs
- NVIDIA CUDA® SDK
- NVIDIA cuDNN, NCCL, and more
- FFTW3 (single, double, and quad precision builds)
- HDF5
- NVBIO
- OpenBLAS
- OpenCV
- Python with H5py, NumPy, pandas, PyCUDA, pydot, scikit-image, scikit-learn, SciPy, SymPy, and more
- MVAPICH2
- OpenMPI
- GNU GCC Compiler Collection (multiple versions, as needed) Provides C, C++ and Fortran compilers.
- Intel OneAPI Cluster Edition (multiple versions, as needed) Provides C, C++ and Fortran Compilers; Integrated Performance Primitives (IPP), Math Kernel Library (MKL), Clik Plus, Threading Building Blocks (TBB), MPI Library, MPI Benchmarks, Trace Analyzer & Collector, VTune Amplifier XE, Inspector XE, Advisor XE
- NVIDIA HPC SDK (multiple versions, as needed) Provides NVIDIA C, C++ and Fortran compilers plus more. Also with GPU-acceleration via CUDA Fortran and OpenACC.
- AMD Optimizing C/C++ Compiler (AOCC) Provides C, C++, and Fortran compilers with optimizations for the latest AMD EPYC CPUs.
