During AMD‘s Financial Analyst Day, the company has launched its detailed plan for the next phase. AMD’s next phase will be driven by multi-generational high-performance CPU and GPU roadmaps and aggressive technology investments.
New roadmaps for data center and computing market
The company plans to unveil a new GPU architecture designed expressly to optimize data center compute workloads, last this year. The new AMD Compute DNA (AMD CDNA) architecture includes second-gen AMD Infinity Architecture to enhance GPU to GPU connectivity and is optimized for machine learning and high-performance computing applications. Secondly, AMD will launch the CDNA 2 architecture. It enables exascale-class supercomputers with its third-gen Infinity Architecture support. Dr. Lisa Su, AMD president and CEO said:
“Our multi-generational computing and graphics roadmaps are designed to significantly accelerate revenue growth and deliver strong shareholder returns. We are focused on relentlessly executing our leadership IP roadmaps and aggressively introducing advanced technologies to drive sustained market share gains across the large and growing high-performance PC, gaming and data center markets.”
Additionally, there are many technology updates, including introducing the first processors based on its next-generation “Zen 3” core in late 2020, 3rd Generation AMD Infinity Architecture with optimized CPU and GPU memory coherence, and more.
At the security side, AMD joined the Confidential Computing Consortium, a group of leading hardware and software companies working to close gaps to protect data through its entire lifecycle to expand its security products portfolio.
Powering El Capitan supercomputer
For the data center market, AMD has delivered significant innovation since 2017, with its first and second-generation AMD EPYC server processors. The company recently announced that El Capitan supercomputer at Lawrence Livermore National Laboratory that delivers more than two exaFLOPs of double-precision performance will be powered. El Capitan supercomputer is planned to become online in 2023.
The introduction of ROCm 4.0, which has been built on previous generations of its ROCm open-source software platform for the data center later this year, is also one of the important parts of the next phase’s plan.