TESLA

  • What is GPU Computing
  • GPU Applications
  • Servers and Workstations
Subscribe
Servers and Workstations

ACCELERATING DISCOVERY AND INSIGHT AT SCALE

The ACCELERATED COMPUTING platform for datacenters

The NVIDIA® Tesla® Accelerated Computing Platform is the leading platform for accelerating big data analytics and scientific computing. The platform combines the world's fastest GPU accelerators, the widely used CUDA® parallel computing model, and a comprehensive ecosystem of software developers, software vendors, and datacenter system OEMs to accelerate discovery and insight. With advanced system management features, accelerated communication technology, and supported by popular infrastructure management software, the Tesla platform provides high performance computing professionals the tools to easily build, test, and deploy accelerated applications in the datacenter.

Accelerate Your Code

Test drive a Tesla K80 GPU
Accelerator >
Accelerate Your Code
 

Accelerating discovery and insight with the Tesla Accelerated Computing Platform

Tesla Accelerated Computing Platform

 

HIGHEST DELIVERED PERFORMANCE

Delivered application performance is more than raw flops. Computational professionals depend on mission critical applications to accelerate discoveries and insights at scale. It starts with the world's fastest accelerators and includes a reliable infrastructure, the ability to monitor and manage that infrastructure, and the ability to quickly move data where it is needed. The NVIDIA Tesla Accelerated Computing Platform provides all of these to deliver unprecedented performance in scientific, analytics, engineering, consumer, and enterprise applications.


The World's Fastest GPU Accelerators

  • Up to 2.91 TFlops double precision performance and Up to 8.74 TFlops single precision performance on the Tesla K80 GPU Accelerator
  • Maximum performance for every application with NVIDIA® GPU Boost™ Technology
  • Increased performance for large data sets with large on-board memory (24 GB on the Tesla K80 GPU Accelerator)
  • Increased throughput to ensure data is available when needed with fast memory bandwidth (480 GB/s on the Tesla K80 GPU Accelerator)
  • Uncompromised data reliability with Error Correcting Codes (ECC) protection for internal GPU memory and ECC plus Dynamic Page Retirement for external GDDR5 memory
Throughput for CPUs compared to Tesla K80 for specific scientific programs and applications.
 

Fast communication

  • Up to 2x faster application performance with NVIDIA NVLINK interconnect technology in future
  • Fast bi-directional communication with the dual DMA engines of Tesla GPU accelerators
  • Accelerated communication across nodes, with network and storage devices, and between GPUs using NVIDA GPUDirect
 

RELIABLE INFRASTRUCTURE Management

  • Easy monitoring and management of GPU clusters with NVIDIA system management tools
  • Enterprise-class InfiniBand, CUDA and OS drivers
  • High performance CUDA driver for Windows OS to reduce CUDA kernel overhead and to enable Windows Remote Desktop and Windows Services
  • Seamless application deployment with ISV Certification for Tesla GPU Accelerators
  • Zero error tolerance stress testing
  • Enterprise level support with three-year warranty for all Tesla products
  • Optional extended support for selected products