Shulou(Shulou.com)11/24 Report--

Thanks to CTOnews.com netizens, South China Daniel Wu, OC_Formula and Prince Guzheng for their clue delivery! CTOnews.com, December 2, AMD CTO Mark Papermaster said at Wells Fargo's sixth TMT Summit in 2022 that the company's next-generation Instinct MI300 accelerated processing unit for data centers and high-performance computing has been up and running in AMD's lab.

AMD Instinct MI300 APU uses AMD's Zen 4 and CDNA 3 architecture, which will be used for El Capitan supercomputing in the United States, which is expected to be put into use in 2024, with the peak computing performance of FP 64 exceeding 2 ExaFLOPS.

AMD's Instinct MI300 is a multi-chip APU with Zen 4 general x86 kernel and CDNA 3-based computing GPU, sharing a unified encapsulated memory pool, including Infinity Cache cache and HBM shared memory design. AMD will use TSMC's N5 (5nm level) manufacturing process to produce CPU / GPU microchips for MI300, but the company did not disclose the number of CPU cores and GPU stream processors in Instinct MI300.

CTOnews.com learned that Papermaster also revealed that AMD is expected to launch EPYC Bergamo processors that compete with Arm servers such as the AWS Graviton series in the first half of 2023, using the Zen 4C architecture.

The AMD EPYC Bergamo processor will have up to 128 cores and will be optimized for HBM-driven Xeon chips and server products from Apple, Amazon and Google, using SP5 slots for higher throughput workloads.

In addition, AMD Genoa-X CPU is expected to go into production by the end of Q3 / early Q1 2023 and will be launched around mid-2023, using a design approach similar to that of Milan-X chips with 3D V-Cache. Genoa-X CPU will be equipped with more L3 cache than 1GB and will be based on the Zen 4 architecture up to 96 cores.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.