Nvidia unveils new technology to accelerate AI, launches new supercomputer

Nvidia’s new Grace CPU Superchip unveiled at the Chipmaker AI Developers Conference was revealed in this undated handout image seen by Reuters. Nvidia / Handout via REUTERS

March 22 (Reuters) – Nvidia Corp (NVDA.O) on Tuesday announced new chips and technologies they said will increase the computing speed of increasingly complex artificial intelligence algorithms, compete against rival chipmakers for lucrative data center business .

The company has provided details of new graphics chips (GPUs) that will be at the core of the AI ​​infrastructure, releasing the H100 chip and a new processor chip called Grace CPU Superchip, based on the technology of the British chip company Arm Ltd. It is the first arm-based chip from Nvidia to be unveiled since its deal to buy arm fell apart last month.

Nvidia has also announced its new supercomputer “Eos”, which it said will be the fastest AI system in the world if it works later this year.

“Data centers become AI factories – processing and refining citizens’ data to produce intelligence,” said Nvidia Chief Executive Officer Jensen Huang at the Nvidia AI Developers Conference online, calling the H100 chip the “engine” of AI infrastructure.

Nvidia said the new technologies together will help reduce computing times from weeks to days for working with training AI models.

Companies have used AI and machine learning for a variety of things – from recommendations for making the next video to watching on TV and cell phones – to new drug discovery.

“It is clear from recent announcements that Nvidia is posing a more significant threat to Intel and AMD in the computing and cloud computing markets,” said Bob O’Donnell, chief analyst at TECHnalysis Research.

Intel Corp. (INTC.O) was the largest manufacturer of core processors for data centers, but increased competition for the lucrative fast-growing space.

However, Vlad Galabov, head of cloud and data center research practice at research firm Omdia, said he had concerns about the power consumption of the H100 chip and said it could hamper the processor’s broad market.

Nvidia’s Chief Financial Officer Colette Kress said that with the new chips pushing AI computing forward, the company’s market potential was about a trillion dollars, from games to chips and systems, and businesses.

Nvidia, whose open-source software was a key driver for companies to use its chips, said it was looking to monetize its software business even more in the future.

“We have already sold software to our companies and this is a few hundred million dollars today and we believe this is a growth opportunity for us,” said Kress, adding that continuing the software business will help Nvidia increase its gross margin at the same time. improve. if chip component shortages and supply constraints have increased costs.

Automotive market software will also be a key driver forward, Huang said. “Car is on track to be our next multi-billion dollar business,” he said.

Nvidia started shipping its autonomous car computer “Drive Orin” this month and Chinese electric car manufacturer BYD Co Ltd (002594.SZ) and luxury electric car manufacturer Lucid Motors (LCID.O) will use Nvidia Drive for their next generation fleets, he said. .

Danny Shapiro, Nvidia’s vice president of automotive, said there would be $ 11 billion in automotive business in the “pipeline” over the next six years, up from $ 8 billion it predicted last year. The growth in previous revenue will come from hardware and from enhanced, recurring revenue from Nvidia Software, Shapiro said.

Nvidia shares closed 0.8% at $ 265.24 on the Nasdaq.

Report by Jane Lanhee Lee, Additional Report by Joseph White; Edited by Bernard Orr

Our Standards: The Thomson Reuters Trust Principles.