NVIDIA Tesla GPU Computing

NVIDIA® Tesla® GPUs deliver supercomputing performance at a lower power, lower cost, and using many fewer servers than standard CPU-only compute systems.

NVIDIA Elite Solution ProviderPowering the world’s leading Supercomputers, Microway designs customized GPU clusters, servers, and WhisperStations based on NVIDIA Tesla and NVIDIA Quadro® GPUs. We have been selected as the vendor of choice for a number of NVIDIA GPU Research Centers, including Carnegie Mellon University, Harvard, Johns Hopkins and Massachusetts General Hospital.

Tesla V100 – World’s Most Advanced Datacenter GPU, for AI & HPC

Integrated in Microway NumberSmasher and OpenPOWER GPU Servers & GPU Clusters

SpecificationsTesla V100 SXM 2.0 GPU

  • Up to 7.5 TFLOPS double- and 15 TFLOPS single-precision floating-point performance
  • NVIDIA Volta™ GPU architecture
  • 5120 CUDA cores, 620 Tensor Cores
  • 16GB of on-die HBM2 GPU memory
  • Memory bandwidth up to 900GB/s
  • NVIDIA NVLink™ or PCI-E x16 Gen3 interface to system
  • Available with enhanced NVLink interface, with 300GB/sec bi-directional bandwidth to the GPU
  • Passive heatsink only, suitable for specially-designed GPU servers

Tesla P100 – Strong Performance and Connectivity for HPC or AI

Integrated in Microway NumberSmasher and OpenPOWER GPU Servers & GPU Clusters

SpecificationsTesla P100 Socketed GPU

  • Up to 5.3 TFLOPS double- and 10.6 TFLOPS single-precision floating-point performance
  • NVIDIA “Pascal” GP100 graphics processing unit (GPU)
  • 3584 CUDA cores
  • 12GB or 16GB of on-die HBM2 CoWoS GPU memory
  • Memory bandwidth up to 732GB/s
  • NVLink or PCI-E x16 Gen3 interface to system
  • Passive heatsink only, suitable for specially-designed GPU servers

Tesla K80 – Density and Performance per Watt

Integrated in Microway NumberSmasher GPU Servers and GPU Clusters

SpecificationsNVIDIA Tesla K80

  • 5.6 TFLOPS single, 1.87 TFLOPS double precision
  • Two GK210 chips on a single PCB
  • 4992 CUDA cores, 2496 per chip
  • 24GB GDDR5 memory (12GB per chip)
  • Memory bandwidth up to 480GB/s
  • Dynamic GPU Boost for performance optimization
  • 8.74 TFLOPS single precision, 2.91 TFLOPS double precision with GPU Boost
  • PCI-E x16 Gen3 interface to system
  • Passive heatsink only, suitable for specially-designed GPU servers

Tesla P40 – Ideal for Deep Learning Inference

Integrated in Microway NumberSmasher GPU Servers and GPU Clusters

SpecificationsPhoto of the front of the NVIDIA Tesla P40 GPU

  • 12 TFLOPS single-precision floating point performance
  • 47 TOPS (tera-operations per second) for inference
  • NVIDIA “Pascal” GP102 graphics processing unit (GPU)
  • 3840 CUDA cores
  • 24GB GDDR5 memory with ECC protection
  • Memory bandwidth up to 346GB/s
  • Dynamic GPU Boost for performance optimization
  • PCI-E x16 Gen3 interface to system
  • Passive heatsink only, suitable for specially-designed GPU servers

Unique features available in the latest NVIDIA GPUs include:

  • NVIDIA GK110 DieHigh-speed, on-die GPU memory
  • NVLink interconnect speeds up data transfers up to 10X over PCI-Express
  • Unified Memory allows applications to directly access the memory of all GPUs and all of system memory
  • Direct CPU-to-GPU NVLink connectivity on OpenPOWER systems supports NVLink transfers between the CPUs and GPUs
  • ECC memory error protection – meets a critical requirement for computing accuracy and reliability in data centers and supercomputing centers.
  • System monitoring features – integrate the GPU subsystem with the host system’s monitoring and management capabilities such as IPMI. IT staff can manage the GPU processors in the computing system with widely-used cluster/grid management tools.

Many of the most popular applications already feature GPU support. Your own applications may take advantage of GPU acceleration through several different avenues:

  • “Drop-in” GPU-accelerated libraries – provide high-speed implementations of the functions your application currently executes on CPUs.
  • OpenACC / OpenMP Compiler directives – allow you to quickly add GPU acceleration to the most performance critical sections of your application while maintaining portability.
  • CUDA integrated with C, C++ or Fortran – provides maximum performance and flexibility for your applications. Third-party language extensions are available for a host of languages, including Java, Mathematica, MATLAB, Perl and Python.

Tesla GPU computing solutions fit seamlessly into your existing workstation or HPC infrastructure enabling you to solve problems orders-of-magnitude faster.

Call a Microway Sales Engineer for Assistance : 508.746.7341 or
Click Here to Request More Information.

Comments are closed.