TL;DR:
- Los Alamos National Laboratory (LANL) is expanding its collaboration with SambaNova Systems to leverage advanced AI hardware.
- SambaNova Systems, backed by significant investments, is making strides in generative AI and large language model (LLM) technologies.
- LANL aims to utilize AI advancements to enhance national security missions.
- The partnership highlights SambaNova’s performance advantage over GPU-based systems in handling complex AI workloads.
- LANL, with a rich history in scientific research, is positioning itself for cutting-edge AI integration.
- SambaNova’s Suite, deployed both on-premises and in the cloud, facilitates the training of generative AI models.
- Accenture and the DOE’s Argonne National Laboratory are among SambaNova’s notable customers.
- SambaNova recently introduced the DataScale SN40L processor, expanding its product offerings.
Main AI News:
In a strategic move aimed at bolstering its capabilities in the realm of artificial intelligence (AI), the Los Alamos National Laboratory (LANL), under the auspices of the US Department of Energy (DOE), is set to expand its partnership with SambaNova Systems. The collaboration will see LANL harnessing the power of SambaNova’s cutting-edge AI hardware solutions.
SambaNova Systems, a San Francisco-based AI technology company that has garnered over $1 billion in investment from industry giants like Google Ventures, Intel, SoftBank, and Singaporean wealth fund GIC, is making significant strides in the field of generative AI and large language model (LLM) technologies. LANL recognizes the potential of these AI advancements in furthering its national security missions, prompting an expansion of its existing partnership.
“The dawn of the exascale supercomputing era demands a deep integration of AI within the ASC computing ecosystem to support our mission objectives today and in the future,” explains Thuc Hoang, program director at NNSA ASC. “We are delighted to scale up our deployment of SambaNova’s Systems, a move that will propel generative AI and LLM technologies to new heights and bolster the ASC program.”
While specific details of the multi-year expansion deal remain confidential, it underscores the confidence that LANL has in SambaNova’s capabilities. Marshall Choy, SVP of product at SambaNova Systems, highlights the company’s edge over GPU-based systems in tackling the most complex foundation model and deep learning workloads. This technological advantage promises to accelerate research, experimentation, and discovery, thereby amplifying the impact of national initiatives.
LANL, situated in Santa Fe, New Mexico, boasts a rich history dating back to its pivotal role in the Manhattan Project during the 1940s. Today, the laboratory houses state-of-the-art supercomputers, including the recently installed Crossroads system.
SambaNova’s Suite, available for on-premises or cloud-based deployment, empowers organizations to train their own generative AI models using popular foundation models. Notably, consulting powerhouse Accenture has already deployed SambaNova’s Suite for generative AI work, further validating its efficacy. The DOE’s Argonne National Laboratory is another esteemed customer in SambaNova’s growing portfolio.
Underlining its commitment to innovation, SambaNova has recently unveiled the DataScale SN40L processor, adding to its suite of offerings. Each DataScale SN30 42U cabinet can accommodate up to three DataScale SN30-8 systems, equipped with eight Cardinal SN30TM RDUs and a total memory capacity of 8TB.
Conclusion:
The expansion of the partnership between Los Alamos National Laboratory and SambaNova Systems underscores the growing importance of AI in advancing national security and scientific research. SambaNova’s continued innovation and performance advantages position it favorably in a market where AI integration is becoming increasingly crucial for organizations with demanding AI workloads.
