Mark Zuckerberg and his wife announce the construction of a high-performance medical research computing system loaded with more than 1000 NVIDIA H100 chips
by
The philanthropic foundation of Mark Zuckerberg and his wife Priscilla Chan has announced that it will fund and build one of the world's largest nonprofit computing systems dedicated to medical research. The system will be equipped with more than 1,000 NVIDIA H100 high-end GPUs with a market price of $40,000 (approximately 5.9 million yen), and will lead to groundbreaking new discoveries that will help treat, prevent, and manage all kinds of diseases. It has been announced that 'It will be.'
CZ Science To Build AI GPU Cluster To Model Cell Systems
https://chanzuckerberg.com/newsroom/czscience-builds-ai-gpu-cluster-predictive-cell-models/
Priscilla Chan & Mark Zuckerberg: How AI can help us understand how cells work—and help cure diseases | MIT Technology Review
https://www.technologyreview.com/2023/09/19/1079261/czi-ai-cell-disease
Machine learning and AI technologies, which have become widely known in the fields of language processing and image generation, are also being used in the life science field, with models such as ' AlphaFold ' and ' ESM ' used to elucidate protein structures, for example. It's coming from technology companies. These models need to process large amounts of data, and can be boosted by adding high-performance GPUs.
AI models built for medical use can be used for a variety of medical research, regardless of protein, but they are typically expensive and may be out of reach for many scientists. .
To provide this model on a non-profit basis, the Chan Zuckerberg Initiative, a philanthropic organization, has announced that it will fund a high-performance computing system consisting of more than 1,000 NVIDIA H100s. This is said to enable the development of new AI models trained on large datasets of cells and biomolecules.
'This system can predict reactions by immune cells, what happens at the cellular level when a child is born with a rare disease, or how a patient's body will respond to a new drug,' said Chan. By scaling up the power of generative AI to biology, researchers can incorporate these technological advances into their studies, accelerating efforts to treat, prevent, and manage all diseases. It will be.”
Although the new system is not as large as those used by private companies for commercial products, once it is operational it will be one of the largest in the world for non-profit scientific research. ``This will be an important resource for academic teams that are hampered by the prohibitive costs of accessing the latest AI technologies,'' said Chan et al.
'Priscilla and I believe that AI will help treat, prevent, and manage every disease this century,' Zuckerberg wrote in a Threads
Demand for AI-related chips is increasing. NVIDIA shipped approximately 300,000 NVIDIA H100 units in the second quarter of 2023, and plans to ship 1.5 million to 2 million units by 2024, but due to manufacturing bottlenecks, the It has also been pointed out that production has been delayed.
TSMC points out that ``semiconductor shortage will continue until around the end of 2024'' - GIGAZINE
by Lee Ji-lin
Related Posts:
