SOUTHFIELD – Wayne State University has received a grant from Silicon Mechanics that provides WSU with a computer research cluster at its Computing Services Center.
The HPC cluster includes hardware and software donated by Intel, NVIDIA, HGST, Mellanox Technologies, Supermicro, Seagate, Kingston Technology, Bright Computing, and LSI Logic. This year?s HPC cluster contains eight compute nodes, one head node, Intel Xeon Phi coprocessors, NVIDIA Tesla GPUs, and InfiniBand and gigabit Ethernet networking.
?We are proud to celebrate with WSU, which stood out in the competitive field of applicants based on the high level of collaboration across departments, as well as the positive impact their research has on faculty, students, and the greater Detroit community,? said Education/Research/Gov’t Vertical Group Manager Art Mann.
The new HPC cluster donated by Silicon Mechanics combines multiple compute and GPU nodes, allowing researchers to best utilize both types of parallel computing. A significant update to WSU?s current computing grid, the cluster will be shared by several of the most computation-intensive research groups on campus, making WSU?s research infrastructure even more competitive.
?The Silicon Mechanics cluster, with the NVIDIA GPUs and Intel co-processors, provides powerful new capabilities for our researchers who are working with highly parallel problems,? said Wayne State?s Associate Vice President and CIO Joe Sawasky. ?This cluster complements our already impressive high performance computing, storage and networking environment, opening the doors to performing cutting-edge research in a fraction of the time it would have taken before.?
Among the groups using the cluster will be two interdisciplinary collaborative research teams, including both computer scientists and domain scientists focusing on chemistry, mathematics, physics, and biology, along with cancer and biomedical research. Fifteen researchers have already begun taking advantage of the cluster?s powerful new parallel processing capabilities.
One research team will use the cluster in its work developing and applying computational chemistry methods for the simulation of chemical and biochemical systems. Another project makes use of parallel computation with multiple GPUs to solve and simulate two-dimensional polycrystalline and/or nanostructured material systems. These systems are important in developing new experimental tools used to understand grain growth dynamics, crucial to the control of hardness and strength in a wide range of engineering materials. Another group will use the cluster?s state-of-the-art hardware, particularly the NVIDIA K40 and Intel PHI co-processor, to enhance the development of WSU?s novel GPU Optimized Monte Carlo simulation engine, known as GO-MC.
