Artificial intelligence (AI) chip maker Habana Labs has announced San Diego Supercomputer Center (SDSC) selected it to provide high-performance inference accelerators and AI training for its Voyager supercomputer.
SDSC’s Voyager, scheduled for release in fall 2021, is earmarked for advancing AI research across a range of science and engineering domains.
"We are honored that Habana's AI processors have been selected to power the AI workloads that will run on San Diego Supercomputer Center's Voyager supercomputer," said Eitan Medina, chief business officer at Habana.
"This implementation of our Gaudi and Goya products showcases how top academic institutions like SDSC can harness efficiency and performance to effectively address the growing demands of AI research workloads."
As part of their deal, SDSC will leverage Habana's interconnect technology to efficiently scale Voyager’s AI training capacity through 336 Gaudi processors. Habana’s Gaudi AI training processors are reportedly the industry's only AI processors that natively integrate ten 100-Gigabit Ethernet ports of RoCE v2 (RDMA over Converged Ethernet) on-chip. The Voyager system will also feature 16 Habana Goya processors to power AI inference models.
Additionally, Habana’s Gaudi AI training and Goya inference processors will enable data scientists and researchers to customize AI models with programmable Tensor Processor Cores and kernel libraries. The models will be implemented through Habana's SynapseAI software platform, which supports popular machine learning frameworks and AI models for applications such as vision, natural language processing, and recommendation systems.
RELATED RESOURCE
IT Pro 20/20: Meet the companies leaving the office for good
The 15th issue of IT Pro 20/20 looks at the nature of operating a business in 2021
According to reports, advanced server solutions provider Supermicro will provide Habana-based AI systems to SDSC for Voyage.
“Combining Supermicro's advanced application-optimized server and storage hardware with Habana's AI training and inference products is the best solution for SDSC's multi-year Voyager AI project," said Ray Pang, vice president of technology and business enablement at Supermicro.
"We continue to work closely with leading innovators to deliver solutions for computationally intensive projects worldwide at leading research and HPC environments for science and medical discovery, compute, and leading-edge AI solutions."
-
The Race Is On for Higher Ed to Adapt: Equity in Hyflex Learning -
Google faces 'first of its kind' class action for search ads overcharging in UKNews Google faces a "first of its kind" £5 billion lawsuit in the UK over accusations it has a monopoly in digital advertising that allows it to overcharge customers.
-
How the National Library of Sweden harnessed AI to unlock centuries of language dataCase Study From Viking-age manuscripts to 1970s broadcasts, AI is helping to digitize more than 18 million items key to Sweden’s history
-
Nvidia and Microsoft team up to build 'most powerful' AI supercomputerNews The firms seek to accelerate the training of AI models using Azure's supercomputer infrastructure and Nvidia GPUs
-
Dell's new PowerEdge servers, quantum solution accelerates HPC for all customersNews The servers were built in partnership with Nvidia and Intel, and are marketed to all organisations - small and large
-
HPE Cray supercomputer to boost Singapore’s met office weather forecastingNews The new system provides twice the speed of its predecessor and has a peak performance of 401.4 teraflops
-
Saudi Arabian university taps HPE to build region’s most powerful supercomputerNews The supercomputer is set to advance insights in areas including clean combustion, Red Sea ecosystems, climate events, and the Arabian tectonic plate
-
HPE's new platform lets customers build machine learning models quickly and at scaleNews The company is also making it easier for businesses to share information from their AI models with each other, without compromising data
-
HPE’s supercomputer helps ISS astronauts experiment in spaceNews Installed in May 2021, HPE claims it’s the first in-space commercial edge computing and AI system to run on the ISS
-
AI-augmented HPC from Dell drives customers successesWhitepaper Solve large data problems faster
