SambaNova Announces Next Generation DataScale System, Setting a World Record for Time-to-Train Performance
SambaNova Achieves Large Language Model Leadership over Nvidia
AI Hardware Summit, Santa Clara, Calif., Sept. 14th, 2022— SambaNova Systems, the company delivering the industry’s only comprehensive software, hardware, and solutions platform to run Artificial Intelligence (AI), deep learning, and foundation model workloads, announces its next generation DataScale® system. DataScale is a fully integrated hardware-software system that enables organizations to reimagine their approach to AI.
As the use of AI continues to accelerate, businesses are looking to move away from tactical AI deployments to scalable, enterprise-wide solutions. SambaNova enables the consolidation of AI sprawl into foundation models that are trained once and reused across the organization.
“The new DataScale SN30 system achieves world record-breaking performance when compared to the latest DGX A100 systems,” said Marshall Choy, SVP of Product at SambaNova Systems. “With this release, SambaNova is also offering subscription pricing for DataScale and Dataflow-as-a-Service™, enabling organizations to achieve ROI faster, reduce risk, and scale more cost-effectively than with any other AI infrastructure.”
The next generation of SambaNova DataScale offers several key enhancements:
- Better Performance: World record GPT training that is 6x faster to train than DGX A100 systems (GPT 13B)
- Next Generation Processor: An enhanced Cardinal SN30 Reconfigurable Dataflow Unit
- Support for the Largest Models: Terabytes of memory — 12.8x more memory capacity than DGX A100
- Enhanced Full Stack Support: Additional deployment automation and ease of scale capabilities
- Customer Value-Based Pricing: Subscription pricing now available for SambaNova DataScale
“We look forward to deploying a larger, multirack system of the next generation of SambaNova’s DataScale systems,” said Bronis de Supinski, Chief Technology Officer (CTO) for Livermore Computing (LC) at Lawrence Livermore National Laboratory (LLNL). “Integration of this solution with traditional clusters throughout our center will enable the technology to have deeper programmatic impact. We anticipate a 2 to 6x performance increase, as the new DataScale system promises to significantly improve overall speed, performance, and productivity.”
“OTP Group has deployed a multi-rack AI system, built on SambaNova Dataflow-as-a-Service GPT, making AI one of our key technology drivers across the region for years to come by utilizing an NLP foundation model,” said Péter Csányi, deputy CEO and head of the digital division at OTP. “The implementation has enabled us to further our commitment to our ESG standards as SambaNova’s solution has enabled the bank to achieve peak AI computing capability in the most energy efficient way.”
“The Argonne Leadership Computing Facility (ALCF) is deploying a multirack system of the next generation of SambaNova’s DataScale systems in the ALCF AI test bed, which provides an infrastructure for the next-generation of AI-accelerator machines,” said Rick Stevens, Associate Laboratory Director and Argonne Distinguished Fellow at the U.S. Department of Energy’s Argonne National Laboratory. “Scientists at Argonne and other research institutions will test this next-generation platform for a variety of use cases including large language models like GPT for gene generation, 3-d convolution networks for neutrino physics, and prediction of tumor response to single and paired drugs. We expect a two to 6x performance increase across a broad spectrum of applications.”
“AI is transforming industries and organizations as dramatically as the internet did,” stated Rodrigo Liang, cofounder, and CEO of SambaNova Systems. “Today’s AI innovators—the banks, energy companies, pharmaceutical companies, and national laboratories— are leaping ahead of competitors by deploying an AI ‘backbone’ that will power the next 10 years of application development and deployment. SambaNova’s solutions, deployed as-a-subscription, will enable them to scale their AI strategies faster and with minimal risk.”
About SambaNova DataScale
DataScale is a fully integrated hardware/software system that can be installed in minutes and enables organizations to train and deploy their own custom state-of-the-art deep learning, foundation model, and deep learning accelerated high-performance computing (HPC) workloads. DataScale is combined with SambaNova’s deep learning expertise to also form the basis of SambaNova’s Dataflow-as-a-Service offering to rapidly analyze, validate, and productize pre-trained foundation models to deliver the greatest value for enterprises and organizations.