Enhanced HPC Architecture and Performance
Today’s high performance computing (HPC) clusters often take advantage of increased node counts, with each node utilizing faster, denser core count processors. With this advancement, scalable performance is a critical component of application optimization on larger, faster clusters. The interconnect is one of the key factors that influences overall performance of the HPC cluster at scale, accounting for up to 30 percent of the cost of an HPC cluster, making price an important consideration.
The Intel® True Scale Fabric architecture was designed from the ground up for HPC, which means it offers improved HPC performance, at a competitive price point, especially for implementations requiring superior performance across large node counts.
Read the full Enhanced HPC Architecture and Performance White Paper.