Nvidia Demonstrates Data Center Prowess at This Year’s European Supercomputer Event
Germany’s ISC Group’s flagship supercomputing event, ISC High Performance 2022, returned to fitness this year in Hamburg, the German Congress Center for the first time in two years. The event, which will last until tomorrow, attracted more than 120 exhibitors. It was there in a special address that Nvidia Corp. drew with thick strokes. his vision accelerated computing using high-performance computing (HPC) solutions powered by its hardware and software.
Nvidia unveiled Los Alamos National Laboratory’s (LANL) new Venado supercomputer, which will deliver 10 exaflops of computing power for artificial intelligence (AI)-related research and development targeting materials science and renewable energy. The company claims that its Arm v9-based Grace Hopper “Superchips” that power Venado will run three times faster than the company’s next-generation graphics processing units (GPUs), the company said. Plus, Grace Central Processing Units (CPUs) can deliver twice the performance per watt of traditional CPUs.
Make the best of a bad situation
Nvidia failed to acquire Arm at the beginning of this year. Regulators in three countries rejected the $40 billion deal, citing its potential anti-competitive effect within the semiconductor industry. SoftBank has chosen to spin off Arm later this year, with an initial public offering (IPO). He will retain a majority stake in Arm, but the news has attracted numerous investors inside and outside the semiconductor business, including Qualcomm and its rivals.
While Arm’s ultimate future may still be in doubt, Nvidia’s faith that Arm opens up new opportunities for accelerated computing is unshakeable. Nvidia founder and CEO Jensen Huang made it clear when the Arm deal fell through that the company’s Arm collaboration from him would continue regardless. Nvidia released Grace Hopper’s designs at his own GTC event a month later.
At last week’s Computex 2022 event in Taipei, Nvidia showed off the first production data center solutions built around Grace Hopper, named for the Navy admiral and computer scientist whose groundbreaking work led to machine-independent programming languages such as COBOL. At the same event, Nvidia introduced liquid-cooled A100 GPUs, designed to reduce data center power use by up to 30% through more efficient cooling.
Building fusion power with Omniverse
While Facebook Meta Maker beats the infrastructure collaboration drum To bring its vision of the metaverse to life, Nvidia is in a very different place. Nvidia’s metaverse equivalent, which he calls the Omniverse, is very real. Omniverse is a collaborative 3D environment used in production for artists, engineers, and research scientists. Ericsson used Omniverse to build simulations of digital twin cities, to aid their 5G small cell densification research ahead of physical deployments. The system helped Ericsson mitigate signal interference and propagation problems, which were previously not possible.
At ISC this week, Nvidia explained that Omniverse is the platform for UK researchers to develop a real-time simulation of a working fusion power plant. Fusion power may not yet exist in the real world, but the researchers hope to greatly reduce the research and development time needed to create a working solution, once the vexing real-world physics and engineering are ironed out.
Accelerating computing research around quantum computing and a new version of Nvidia’s imaging software, Holoscan, targeting edge HPC settings rounded out Nvdia’s ISC 2022 announcements.