Eviden introduces new AI computing portfolio

 This full end-to-end offering includes five new AI server ranges optimized for AI workloads and is designed to cater to a diverse range of AI needs. This includes inferencing AI models at the edge, fine-tuning existing models in enterprise data centers, or training foundation models with AI supercomputers. Each model offers varying levels of performance, scalability, and flexibility, and can be selected based on each specific use case, budget, and business size.

This offering leverages use-case focused AI software, including Eviden’s Computer Vision VISuite and global AI services (specialized consulting services, deployment, orchestration, industrialization services, and support) provided by Eviden’s global data scientist team to help clients tackle specific industry challenges. The BullSequana AI product line will also offer NVIDIA AI Enterprise, an end-to-end cloud-native software platform that accelerates data science pipelines and streamlines the development and deployment of production-grade copilots and other generative AI applications.

BullSequana AI 1200H Flagship Product

As part of this new range, Eviden presents its flagship product: the BullSequana AI 1200H, which combines cutting-edge technologies in AI acceleration and high-density computing with specialized consulting services and is designed to support businesses in achieving their goals more energy-efficiently.

The first customer for this new offering is GENCI (Grand Equipment National de Calcul Intensif) and CNRS (Centre National de la Recherche Scientifique) in France which have selected the BullSequana AI 1200H with 1456 NVIDIA H100 GPUs to extend the capacity of their Jean Zay supercomputer,  increasing its peak computing power from 36.85 to 125.9 Pflop/s. Operated by CNRS’s IDRIS (Institut de développement et de ressources en informatique scientifique), it will be fully available to users early this summer in order to answer to a high demand on generative AI.

The BullSequana AI 1200H is an enterprise-grade supercomputer specifically designed for use cases requiring high computational power for training, fine-tuning, and running inference of large language models (LLMs) and for the complex systems modeling required by so many industries and applications today. These applications demand a considerable amount of computing power due to their large capacity for deep learning tasks with many different parameters and complex architecture. By leveraging cutting-edge hardware and software technologies, Eviden provides highly optimized solutions customized to meet each client’s specific needs. Its team of experts provides the highest level of service and works closely with its clients to ensure they have the support and guidance they need to succeed.

Best-in-Class Energy Efficiency and Density

This offering, specifically designed for AI, is among the most energy-efficient and dense solutions available on the market. Because of Eviden’s patented Direct Liquid Cooling (DLC) technology, Eviden BullSequana AI supercomputers are at least 3 times denser than any traditional air-cooled systems, and additional heat can be transformed into energy to re-heat nearby buildings. This solution is built and produced at Eviden’s factory in Angers, France, allowing businesses to meet their data sovereignty and security objectives.

AI-Powered

Eviden is collaborating with NVIDIA to integrate some of the most powerful components in its product portfolio into this new offering, including NVIDIA GH200 Grace Hopper Superchips. Combined with Eviden’s Smart Energy and Management Suite, this new offering is well-positioned to speed up companies’ AI adoption and help them harness the power of AI. Moving forward, Eviden plans to integrate NVIDIA GB200 Grace Blackwell Superchips.

Dr. Cédric Bourrasset, Head of Quantum & AI Computing, Eviden, Atos Group, said, “AI is a transforming force capable of revolutionizing businesses across industries and sectors by enhancing productivity and accelerating innovation. However, developing AI applications requires a great deal of expertise, high computing power, and energy to design, train, and run AI models. This is particularly true for foundation models. With Eviden’s new offering, customers who require reliability, scalability, and versatility to stay ahead of innovation will have maximized acceleration and the right expertise in AI to enable them to make their ideas and innovations a reality.”

“Eviden’s end-to-end solution, capitalizing on its expertise with the most renowned French and worldwide AI computing research centers, offers the most complete AI computing platform to help customers accelerate time to AI operationalization and industrialization of innovative applications in areas from healthcare, new energies, material design to legal or mass distribution” said Stephane Requana, CTO, GENCI.

“Generative AI is transforming every industry, and its development demands expertise and sufficient AI computing resources,” said Serge Palaric, Vice President of Alliances and EMEA OEMs at NVIDIA. “Powered by the NVIDIA Grace Hopper platform, Eviden’s new BullSequana AI portfolio, through collaboration with renowned research centers, accelerates the deployment of innovative applications across industries.”

BullSequana AI 1200H Configuration

Hardware: First BullSequana AI 1200H DLC configuration is equipped with NVIDIA Grace Hopper Superchips* and NVIDIA Quantum-2 InfiniBand networking. Software: Eviden Jarvice™ XE, Eviden Smart Energy Management Suite, Eviden Smart Management Center, and NVIDIA AI Enterprise.