SiPearl partners with Samsung for built-in HBM in Rhea
SiPearl, the company building the high-performance low-power European microprocessor for HPC and AI inference, has signed a partnership with Samsung to equip its Rhea series with the South Korean company's memory solution.
Europe consumes nearly 30% of the world’s supercomputing resources, only 5% of supercomputers are supplied in Europe and exactly 0% of the microprocessors powering them are designed here. This is something that the European Processor Initiative (EPI) is looking to change.
The European Processor Initiative is the EU's response to this strategic autonomy challenge. SiPearl is a private company which was created within the EPI. Located in France, Germany, and Spain, SiPearl designs the next energy-efficient microprocessor for HPC – and the subsequent generations of SiPearl microprocessors will power Europe’s data centres and cloud infrastructures.
After an extensive market survey, SiPearl has selected Samsung to equip its Rhea series with the South Korean company's High-Bandwidth Memory (HBM).
“We are excited to contribute to the expansion of SiPearl’s unique supercomputer/AI ecosystem with Samsung’s advanced HBM technology,” says Yongcheol Bae, Executive Vice President and Head of Memory Product Planning at Samsung Electronics, in a press release. “As the HPC and AI market grows and diverse requirements increase, high-performance, high-bandwidth, low-power solutions are key to support development of the Rhea series in the future. We are committed to delivering memory solutions that can bring superior performance and efficiency for future HPC and AI applications in a close collaboration with SiPearl.”
“At SiPearl, we are thrilled to be a European pioneer in the use of HBM technology in conjunction with Samsung Electronics. Combining our high-performance, low-power microprocessor technologies with built-in HBM is key to meet all the requirements of supercomputing and AI inference workloads. On the fast-growing AI inference market, we think that our product will be a market leading solution for inference Large Language Model (LLM) tasks providing among other things high flexibility to model changes relative to solutions currently in use ,” says Philippe Notton, CEO and founder of SiPearl.
