Related News

AMD, Intel and Nvidia present new accelerators in SC12

Advanced Micro Devices (AMD), Intel and Nvidia have presented their new compute accelerators cards for high-performance computing (HPC) at the SuperComputing Conference & Exhibition (SC12), taking place in Salt Lake City (USA).

The new products deliver better theoretical performance compared to previous generation without increasing the energy consumption. To achieve this the level of parallelism (internal computing cores) is greatly increased.

AMD's FirePro S10000 compute accelerator card is designed for both high-performance computing (HPC) and workstations, with 6GB of GDDR5 memory with ECC and delivers 5.91 TFLOPS of peak single-precision and 1.48 TFLOPS of double-precision floating-point performance. It is based on the Graphics Core Next Architecture.

Nvidia Corp. introduced its Tesla K20-series of graphics processing units-based accelerators for high-performance computing (HPC) applications. The new products are based on the code-named GK110 processors specifically tailored for the HPC world. Nvidia K20X with 2688 stream processors and 6GB of GDDR5 memory offers 3.95TFLOPS single-precision and 1.31TFLOPS double-precision peak floating point performance.

Intel Corp. on Monday introduced the first commercial Xeon Phi coprocessors based on Intel many-core MIC micro-architecture for the high-performance computing market. The world’s largest chip supplier has already started to ship Xeon Phi products to customers on the market of supercomputers. The Intel Xeon Phi coprocessor 5110P features 60 cores with 4-way simultaneous multi-threading technology and 512KB L2 cache per core, provides additional performance at a lower power envelope. It reaches 1.01TFLOPS double-precision performance, and supports 8GB of GDDR5 memory at a higher 320 GB/sec memory bandwidth.

More information about AMD FirePro S10000

More information about Intel Phi

More information about Nvidia K20-series