NVIDIA Has Unveiled The Tesla K40 GPU Accelerator

Posted Nov 18, 2013

NVIDIA has announced the Tesla K40 GPU accelerator today.  The Tesla K40 is the world’s highest performance accelerator ever built.  The Tesla K40 supports extreme performance to a wide range of high performance computing and enterprise applications.  It can provide double the memory and up to 40% higher than the Tesla K20X GPU accelerator and 10 times the higher performance than today’s fastest CPU.


The Tesla K40 is the world’s first and highest performance accelerator optimized for big data analytics and large-scale workloads.  The Tesla K40 has intelligent NVIDIA GPU Boost technology, which converts power headroom into a user-controlled performance boost.

“GPU accelerators have gone mainstream in the HPC and supercomputing industries, enabling engineers and researchers to consistently drive innovation and scientific discovery,” stated Sumit Gupta, general manager of Tesla Accelerated Computing products at NVIDIA. “With the breakthrough performance and higher memory capacity of the Tesla K40 GPU, enterprise customers can quickly crunch through massive volumes of data generated by their big data analytics applications.”

The Tesla K40 surpasses other accelerators on two common measures of computational performance: 1.) 4.29 teraflops single-precision and 2.) 1.43 teraflops double-precision peak floating point performance.

Some of the other features of the Tesla K40 GPU accelerator includes:

1.) 12GB of ultra-fast GDDR5 memory lets users process 2X larger datasets, which enables them to quickly analyze massive volumes of data.

2.) 2,880 CUDA parallel processing cores deliver application acceleration by up to 10X compared to using a CPU alone.

3.) Dynamic Parallelism enables GPU threads to dynamically spawn new threads, enabling users to quickly and easily crunch through adaptive and dynamic data structures.

4.) PCIe Gen-3 interconnect support accelerates data movement by 2X compared to PCIe Gen-2 technology.

“The Tesla K40 GPU accelerators will help researchers crunch through massive volumes of big data and gain new insights through large-scale, sophisticated visualizations,” said Kelly Gaither, director of Visualization at the Texas Advanced Computing Center (TACC). “With NVIDIA GPUs, Maverick will provide researchers powerful interactive capabilities to advance their most complex scientific challenges.”

The TACC plans to launch a new project called Maverick, which is a remote visualization and data analysis system powered by the NVIDIA Tesla K40 GPU accelerators.