Nvidia Unveils New Data Center Chips to Speed Pace of AI

The new Ampere chip design is 20 times faster than its predecessors and can be repurposed for both training and inference.

Bloomberg

May 14, 2020

2 Min Read
Nvidia's A100 Ampere GPU
Nvidia's A100 Ampere GPUNvidia

Ian King (Bloomberg) -- Nvidia Corp. announced new graphics chips and computers for artificial intelligence processing in data centers, claiming a leap forward in performance that may help cement its lead in the growing area.

The company’s Ampere chip design is 20 times faster than its predecessors and has the flexibility to be repurposed between the two key areas of artificial intelligence processing: training and inference, Chief Executive Officer Jensen Huang said. The company unveiled the new chip, which Huang called the biggest leap in Nvidia’s history, Thursday at an online technology conference.

Nvidia’s success in building a $3 billion business over the last five years has attracted the attention of rivals and startups trying to offer novel solutions for AI processing. Nvidia, the biggest maker of graphics chips used in gaming personal computers, has adapted its technology for AI. Huang’s products show the value of systems that manipulate data in parallel, making it possible to distill patterns from the massive flood of information in an increasingly connected world.

Ampere chips are already part of servers shipped by Nvidia. The company is pitching them as being capable of replacing much bigger, more expensive and power-hungry racks of machinery that feature a lot of Intel Corp. processors.

Underling the cost of such systems, Nvidia is selling Ampere-based computers called DGX for $199,000 each. In a typical data center, five of those machines would outperform 600 servers based on Intel chips and the previous generation of Nvidia products, an array that would have cost $11 million, according to Nvidia.

Nvidia chips have grabbed market share in the AI area of training, which is the rapid crunching of huge amounts of data -- such as images or sound files -- to determine patterns that are then embedded into software. Inference, running that resulting software to provide real-time responses for things like voice commands, has been mainly done on servers that use Intel Xeon chips.

Huang’s latest pitch is that the new Ampere design can tackle both tasks and switch between the two types of work as needed. While Nvidia is introducing the new product, Intel is about to debut a graphics chip-based processor design aimed at taking Nvidia head on. Advanced Micro Devices Inc., Nvidia’s biggest rival in computer gaming graphics chips, is also touting new capabilities for data centers.

Related:How Cumulus Fits into Nvidia’s Data Center Play

Nvidia and its competitors are going after the giant capital spending budgets of the public cloud-computing companies such as Amazon.com Inc., Alphabet Inc.’s Google and Microsoft Corp. Huang declined to discuss specific customers, but said that all of the major cloud providers have already taken delivery of Ampere-based computers.

Related:Nvidia and Microsoft Azure Pitch Rentable Cloud Supercomputers

Subscribe to the Data Center Knowledge Newsletter
Get analysis and expert insight on the latest in data center business and technology delivered to your inbox daily.

You May Also Like