sales@microway.com
Phone: 508-746-7341
Fax: 508-746-4678
Microway (R) Technology you can count on.

NVIDIA Tesla




 
NVIDIA GPU Test Drive Program
GPU Test Cluster
Benchmark AMBER, NAMD and more on Tesla K20!


Greener HPC with Microway and Tesla Kepler
Clusters, Servers, and WhisperStations™ with Tesla K10 and K20 GPU Accelerators

Microway HPC Clusters and WhisperStations seamlessly integrate the NVIDIA® Tesla™ GPU compute processors. Solve your most important HPC challenges with a massively multi-threaded, parallel-computing architecture, and the CUDA™ C, C++, and Fortran language environment and tool suites.

The lastest Tesla K10 and K20/K20X GPUs provide the most cutting edge floating point performance and must-have HPC features. Solutions are available for all scales and budgets.
  • Massively parallel SMX architecture with over 3X the performance per watt of Tesla Fermi GPUs
    • 2496 SMX CUDA cores in Tesla K20
    • 2688 SMX CUDA cores in Tesla K20X
    • Up to 1/20th the power consumption and 1/10th the cost of equivalent CPU solutions
  • Innovative "Kepler" architecture simplifies code development and increases computational efficiency
    • Dynamic Parallelism allows GPU code to operate without intervention from the CPU
    • Hyper-Q enables multiple applications to simultaneously access the GPU (up to 32 at a time)
  • Up to 3.95 TFLOPS single- and 1.33 TFLOPS double-precision floating point throughput
    • NVIDIA GK110 GPU provides IEEE 754 single & double precision floating point units
  • Up to 6GB high-speed GDDR5 memory with ECC
    • 5GB GDDR5 with 208 GB/sec bandwidth on Tesla K20
    • 6GB GDDR5 with 250 GB/sec bandwidth on Tesla K20X
  • Accessible and flexible programming environments
    • Native NVIDIA CUDA C/C++ environment with profiler and debugger
    • PGI Accelerator with OpenACC for C, C++ and Fortran
    • MathWorks PCT for MATLAB on GPUs
    • Common libraries (BLAS, FFT, RAND, SPARSE, Thrust, CULA) may be called for instant GPU acceleration
    • OpenCL and DirectCompute capable
  • Designed for integration in HPC environments
    • Actively cooled Tesla K20 GPUs pair with new or existing workstations and servers
    • Passively cooled Tesla K10, Tesla K20, and Tesla K20X GPUs may be integrated into specially-designed GPU servers and clusters

Workstations

WhisperStation- Tesla K20

WhisperStation Maximus
Clusters

Tesla GPU Clusters

S2075/S2090 GPU Servers

Microway BioStack-LS
Servers

NumberSmasher/Navion-1U GPU

NumberSmasher- 2U GPU

NumberSmasher- 4U GPU

Microway OctoPuter

*With ECC enabled, memory available to the user will be 5.25GB per GPU for K20X
Privacy    Legal Our mission is to provide customers with leading edge technologies for high performance computing solutions (HPC). We establish and maintain industry recognized products and expertise for beowolf cluster interconnect (InfiniBand), clustre management and HPC storage solutions.