Intel announces new coprocessor, Nvidia releases new K20 GPU
Our new Pinnacle Flex system, described above, can accommodate either of the two newest products on the market: the Intel Xeon Phi coprocessor and the Nvidia Tesla K20 GPU.
At SC12, Intel introduced its Xeon Phi coprocessor, which optimizes highly parallel computing with its many-integrated-core (MIC) architecture. This coprocessor plugs into a PCIe slot and features more than 50 cores, working in tandem with traditional processors to deliver high memory bandwidth and floating point performance.
The Xeon Phi is Intel's first MIC product, and two models will be produced in the first half 2013. The Xeon Phi 5110P, optimized for high density environments, is scheduled to start shipping on January 28. This 1.05GHz model features a peak memory bandwidth of 320GB/s for a double-precision floating point performance of just over one teraflop.
The Xeon Phi 3100-series coprocessor, designed for highly parallel, compute-bound workloads, will be available by the end of June.
You can view additional details about the Intel Xeon Phi here: www.advancedclustering.com/products/xeon-phi.html.
Recently made available on the market is Nvidia's new Tesla K20 GPU. Unlike the previously released K10, which is designed for single-precision computing, the K20 is optimized for parallel-computing with two new performance-doubling features: Hyper-Q and Dynamic Parallelism.
The Hyper-Q feature allows the GPU to tackle up to 32 message passing interface (MPI) processes simultaneously, as opposed to earlier GPU models that could only handle one MPI process at a time. Dynamic Parallelism allows the GPU to distribute work among its cores rather than sending requests back to the CPU each time subsequent calculations need to be made.
For more information on Nvidia's Tesla K20 GPUs, please visit Nvidia's blog posts:
If you have any questions about these new products and how we can integrate them into systems designed for your needs, please contact us at