AMD partners with Arm developers for exascale computing

AMD has announced plans to partner with French chip designer SiPearl to build an exascale supercomputing system that uses SiPearl’s Arm-based Rhea processors and AMD’s Instinct GPU accelerators.

SiPearl is a relatively small startup that began operating in 2019, licensing Arm’s Neoverse high-performance technology. It has many alliances with partners including Intel, Nvidia, HPE and Graphcore.

SiPearl also participates in the European Processor Initiative (EPI), a consortium selected by the European Union to support the development of microprocessors in Europe specifically for emerging applications such as high-performance computing (HPC) and artificial intelligence. EPI aims to develop Arm-based processors for exascale supercomputers by 2023.

SiPearl’s Rhea microprocessor was designed in partnership with French IT giant Atos. It comes with 72 next-generation Arm Neoverse cores and supports DDR5 and high-bandwidth HBM2E memory interfaces. This gives it extremely fast memory (HBM) and a very large memory pool (DDR5). The processor will be manufactured by TSMC and is expected to ship in 2023.

Instinct accelerators are based on AMD’s Radeon GPU technology. They can be found in other systems, including Frontier, Oak Ridge National Laboratory’s exascale supercomputer, which is the fastest supercomputer in the world.

AMD and SiPearl said they will first begin studying the interoperability of AMD’s ROCm software stack with SiPearl’s Rhea microprocessors. ROCm is AMD’s software stack for developing GPU-accelerated applications. This is a form of future-proof software ecosystem, so future applications will run on supercomputers developed by both companies.

The joint effort will first focus on porting and optimizing AMD’s Heterogeneous Portability Interface (HIP), AMD’s GPU programming environment that allows developers to write code that will run on AMD and Nvidia GPUs.

Philippe Notton, CEO and founder of SiPearl, said in a statement: “It provides European supercomputer end-users with a wider choice and will enable Europe to tackle the great challenges of our time, such as artificial intelligence, climate change and climate change.” modeling and medical research.”