Network Security Internet Technology Development Database Servers Mobile Phone Android Software Apple Software Computer Software News IT Information

In addition to Weibo, there is also WeChat

Please pay attention

WeChat public account

Shulou

AI strategy is not only Instinct MI300A,AMD EPYC, but also an important part.

2024-04-23 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > IT Information >

Share

Shulou(Shulou.com)12/24 Report--

CTOnews.com12 news on March 7, yesterday, AMD held a theme event called "Advancing AI" in San Jose, California, and officially launched a new generation of APU--Instinct MI300X / MI300A for AI / HPC, the former is the traditional GPU accelerator, while the latter is the fusion accelerator of CPU+GPU. We can still see a lot of EPYC in it. For example, the CPU part of the MI300A fusion accelerator directly reuses the design of EPYC. In addition to integrating six XCD GPU modules, four IOD modules, and eight HBM3 memory modules, MI300A also has three CCD modules, which are the same.

At that time, AMD's EPYC processor also had powerful AI capabilities. At the data Center and Al Technology Premiere in June, AMD optimized for cloud native load and launched a range of new products, including the AMD EPYC 9704 processor "Bergamo", which is throughput-oriented and has the highest terminal density and efficiency, and Genoa-X for technical computing.

Among them, the highlights of Genoa-X include leading 5nm process node technology, up to 96 high-performance Zen4 cores, L3 Cache up to 1.1GB, compatible SP5 slots, and so on. These features enable Genoa-X to accomplish more design tasks every day in Ansys CFX, thus greatly accelerating product development.

This makes AMD's EPYC processor and Instinct accelerator the preferred solution for many of the world's leading supercomputers, whether in consumer areas or in professional areas such as data centers. According to the TOP500 list of the 62nd global supercomputer ranking, 140of the 500supercomputing platforms are from AMD.

Since its inception in 2017, AMD EPYC has evolved for four consecutive generations, demonstrating the highest computing density, excellent performance and efficiency in the industry. It has many cores, huge cache, high frequency and rich technical features. At the same time, it also has a very high performance-to-price ratio, which is undoubtedly the first choice for data center customers.

At the meeting, another APU product, MI300A, integrates Zen3 CPU and CDNA3 GPU, and uniformly uses HBM3 memory and the fourth-generation Infinity Fabric high-speed bus interconnection, thus realizing structural simplification and programming convenience. With the support of the unified architecture, the APU accelerator can give full play to its performance potential.

At the same time, AMD also unveiled the Ruilong 8040 series processor in the latest generation of notebook products, which inherits the Ryzen AI NPU engine and increases its speed by up to 60%. The Ryen 8050 APU, code-named Strix Point, will be equipped with a XDNA 2 engine, and in particular, the AI performance improvement will be more than tripled.

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.

Views: 0

*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.

Share To

IT Information

Wechat

© 2024 shulou.com SLNews company. All rights reserved.

12
Report