Intel Max Series Brings Revolutionary Memory Bandwidth & Performance to HPC and AI

Intel Corporation has unveiled the Intel Max Series product family in time for Supercomputing '22 in Dallas, featuring two cutting-edge items for high-performance computing (HPC) and artificial intelligence (AI). The Intel® Xeon® CPU Max Series (code-named Sapphire Rapids HBM) and the Intel® Data Center GPU Max Series (code-named Ponte Vecchio). Updates on the deployment of the new products, which will power the planned Aurora supercomputer at Argonne National Laboratory, were released today. The Xeon Max CPU accelerates numerous HPC applications without requiring any code changes because it is the first and only x86-based processor with high bandwidth memory.  [caption id="attachment_61385" align="aligncenter" width="1200"] Image Credit: Intel[/caption] With up to 128 gigabytes (GB) of high bandwidth memory and over 100 billion transistors crammed into a 47-tile box, the Max Series GPU is Intel's dense CPU. For these new CPUs, a single development environment is offered through the one API open software ecosystem. The advanced features of the Intel Max Series products will be enabled via Intel's one API and AI tools in 2023. What's Important: High-performance computing (HPC) is at the cutting edge of technology, utilizing the most cutting-edge advancements at scale to address the most pressing issues in science and society, from reducing the effects of climate change to finding cures for the most lethal diseases. [caption id="attachment_61386" align="aligncenter" width="480"] Image Credit: Intel[/caption] With scalable, balanced CPUs and GPUs that incorporate memory bandwidth advancements and are connected by one API, an open, standards-based, cross-architecture programming framework, the Max Series solutions satisfy the needs of this community.  Using devices from the Max Series, researchers and enterprises may address challenges more quickly and sustainably. When It Will Arrive: In January 2023, the Max Series items are expected to go on sale. As promised to customers, Intel is delivering Xeon Max CPUs to Los Alamos National Laboratory, Kyoto University, and other supercomputing centers in addition to supplying blades with Max Series GPUs to Argonne National Laboratory to run the Aurora supercomputer.