Two major themes are defining today’s HPC era: a global race toward exascale-class computing, and simultaneously, a convergence between HPC, big data and artificial intelligence (AI). Intel reports that it plans to deliver the foundational elements of both trends. For example, Intel reports that it is enabling scalable performance for a broad range of HPC systems from the smallest clusters to the world’s largest supercomputers, while processing various AI and big data workloads.
On the exascale front, Intel has been collaborating with many global academic and national research institutions on their HPC needs for the path to exascale. One effort has been with the U.S. Argonne National Labs as part of the CORAL program. Recently, the Department of Energy announced it intends to deploy the first U.S. exascale system based on Intel technology. To meet customers’ needs in exascale, Intel has shifted the focus of collaboration toward a new platform and microarchitecture specifically designed for HPC, big data and AI at the highest scale.
Regarding HPC, Intel announced the addition of a new family of HPC solutions to the Intel Select Solutions program. Built on the latest Intel Xeon Scalable platforms, Intel Select Solutions are verified configurations designed to speed and simplify the evaluation and deployment of data center infrastructure while meeting a high-performance threshold.
Intel announced the addition of the Intel Select Solutions for the HPC market, including Intel Select Solutions for Simulation and Modeling, aimed to help more companies accelerate product designs that offer customization and optimization.
OEMs will soon begin offering platforms based on Intel Select Solutions for HPC from global solution providers such as Atipa, Boston Ltd., Colfax, E4 Computer Engineering, Exxact, Inspur, Lenovo, Megware, Nortech, RSC and Supermicro.
Intel announced a new 48-port leaf module with a double-density connector for dense Intel Omni-Path 100 series director switch solutions. With the 48-port module offering, the existing director switch chassis can now support 50% more ports, or up to 1,152 ports in the 20U chassis. This helps reduce the number of switches to help users run their HPC and deep learning jobs faster. The 48-port leaf module will be available early in 2018.
For more info, visit Intel.
Sources: Press materials received from the company.