On January 10, Intel will formally launch the 4th Generation Intel® Xeon® Scalable CPUs and the Intel Xeon CPU Max Series, as well as the Intel Data Center GPU Max Series for high performance computing (HPC) and artificial intelligence (AI).
The event will highlight the value of 4th Gen Intel Xeon Scalable processors and the Intel Max Series product family, as well as customer, partner, and ecosystem support, and will be hosted by Sandra Rivera, executive vice president and general manager of Intel’s Datacenter and AI Group, and Lisa Spelman, corporate vice president and general manager of Intel Xeon Products.
The event will show how Intel is addressing critical market needs with a workload-first approach, performance leadership in key areas such as AI, networking, and HPC, the advantages of security and sustainability, and how the company is delivering significant outcomes for its customers and the industry.
The Intel ‘Ponte Vecchio’ GPU, or the ‘Intel Data Center GPU Max Series,’ as it is currently known, is a significant product with 128 Xe Cores, 128 RT Cores (making it the only HPC / AI GPU with a native raytracing core), up to 64 MB of L1 cache, and up to 408 MB of L2 cache.
There is also 128GB of HBM2e utilised, and the IO can connect up to 8 discrete dies. PCIe Gen 5 is utilised in conjunction with Xe Link to give massive computing power. It is built using a combination of Intel 7, TSMC N5 and TSMC N7 chips packaged utilising the EMIB and Foveros methods.
Intel Sapphire Rapids Xeon Max CPUs with HBM memory will support up to four HBM packages, each with much better DRAM bandwidth than a Sapphire Rapids-SP Xeon CPU with 8-channel DDR5 memory. This will enable Intel to offer a chip with enhanced capacity and bandwidth to customers who request it. The HBM SKUs can be utilised in two modes: HBM Flat and HBM caching.
Also Read:
