Intel teams with Google Cloud to optimize HPC performance

Intel and Google Cloud have co-developed a new toolkit to simplify and optimize high-performance computing (HPC) workloads on Google Cloud.

Dubbed Cloud HPC Toolkit, the new platform allows enterprise customers to create and replicate turn-key HPC clusters. Backed by Intel’s 2nd Gen Xeon Scalable processors, the solution also features the firm’s Select Solutions.

Additionally, the new resource provides access to tools from “Intel oneAPI Base and HPC Toolkits to optimize performance through Intel Select Solutions for Simulations & Modeling.”

“Using Cloud HPC Toolkit with an Intel Select Solutions for Simulations & Modeling blueprint brings the added benefit of automatically spinning up a hardware-software configuration that has been rigorously tested and optimized for real-world performance, eliminating guesswork,” stated Ilias Katsardis, HPC solution lead at Google.

Furthermore, built-in Intel oneAPI tools streamline HPC programming by offering support for C++, SYCL, Fortran, OpenMP, MPI, and Python. Commenting on integrations, Google affirmed support for Altair PBS Professional and Altair Grid Engine job scheduler in later versions.

Intel had previously partnered with Google to automate the creation of Intel Select Solution-compliant HPC clusters on Google Cloud. The recent announcement reinforces the duo’s longstanding partnership.

“DAOS is the future of HPC. We’re excited to see the benefits of our year plus technical collaboration pay off with today’s announcement of fast and easy access to DAOS in Google Cloud,” commented Kelsey Prantis, director of engineering, high-performance storage, at Intel.