The new products will be the basis for Argonne National Laboratory’s upcoming Aurora supercomputer
Intel announced the Intel Max Series product family, which includes two products for High Performance Computing (HPC) and artificial intelligence (AI): Intel Xeon CPU Max Series (formerly known as “Sapphire Rapids HBM”) and Intel Data Center GPU Max Series (codenamed “Ponte Vecchio”). The new products will be the basis for Argonne National Laboratory’s upcoming Aurora supercomputer. The Xeon Max CPU is the first and only x86 processor with High Bandwidth Memory (HMB) and will allow you to accelerate many HPC workloads without requiring code modification. The Max Series GPU is Intel’s highest density processor, with over 100 billion transistors in a 47-tile package with up to 128 gigabytes (GB) of High Bandwidth Memory. The oneAPI open software ecosystem provides programmers with a single environment for both new processors. In 2023, Intel’s oneAPI and AI tools will offer features that will allow you to take advantage of the capabilities of Intel Max Series products.
“To ensure that no HPC workload is too heavy, you need a solution that maximizes bandwidth, computing power and developer productivity. The Intel Max Series product family leads to a broader High Bandwidth Memory and oneAPI market, making it easier to share code between CPU and GPU and solve key challenges in this area faster, ”said Jeff McVeigh, Intel Corporate Vice President & General Manager, Super Compute Group. The Max Series product launch is scheduled for January 2023. Intel is already shipping expansion modules with Max Series GPUs to Argonne National Laboratory for use in its Aurora supercomputer and will deliver Xeon Max CPUs to Los Alamos National Laboratory, Kyoto University and other supercomputer sites. High Performance Computing (HPC) is at the forefront of technology and uses the most advanced innovations to respond to scientific and social challenges, from finding solutions to respond to climate change to treating the deadliest diseases.
The Aurora supercomputer, currently under construction at Argonne National Laboratory, is expected to be the first supercomputer to exceed 2 peak exaflops in double-precision computational performance in 2023. Aurora will also be the first to showcase the power made available by the combination of Max Series CPUs and GPUs in a single system, with over 10,000 blade modules, each of which will contain six Max Series GPUs and two Xeon Max CPUs. Argonne and Intel unveiled Sunspot, Aurora’s Test Development system, with 128 blade modules. Aurora Early Science Program researchers will have access to the system from the end of 2022. Max Series products will be used for several other HPC systems critical to national safety and scientific research, including Crossroads at Los Alamos National Laboratory, CTS-2 systems at Lawrence Livermore National Laboratory and at Sandia National Laboratory and Camphor at Kyoto University.