Nvidia bags Amazon Web Services in its latest data-center chip push


Nvidia Corp. attempted to show progress for its newest chips Monday, as Chief Executive Jensen Huang kicked off the company’s annual hometown conference.

At the keynote address of the 2019 GTC Conference in San Jose, Calif., Huang detailed new developments for the Turing-based equipment that Nvidia














NVDA, -0.51%












 has rolled out in the past year. Most important was Nvidia’s announcement that Amazon.com Inc.














AMZN, +1.74%












 will now be using its T4 data-center chips, the biggest name that had not yet publicly signed on with Nvidia.

The Santa Clara, Calif.,-based chip maker said its T4 Tensor Core graphics processing units, or GPUs, would be deployed to Amazon Web Services through Elastic Compute Cloud G4 in the coming weeks. While other public cloud services have been chipping away at market share over the past few years, Amazon’s AWS still ranks as a global market-share leader in public cloud services.

“If you want to reach a lot of people, and you want to reach a lot of people fast, with the single largest compute engine on the planet, there’s one way of doing it,” Huang said in announcing the AWS deal as the keynote address stretched past two hours long.

Back in September, Nvidia announced that Alphabet Inc.’s














GOOG, -0.02%













GOOGL, -0.15%












 Google Cloud Platform would use T4 chips in its data centers. In addition to Google, Ian Buck, Nvidia’s general manager of accelerated computing, said in a briefing that Baidu Inc.














BIDU, +0.40%












  had adopted T4 chips for its data centers, and that Alibaba Group Holding Ltd.














BABA, +0.48%












 also planned to announce adoption of T4 chips, with the last notable holdout being Microsoft Corp.’s














MSFT, +1.43%












 Azure cloud service.

Nvidia has been seeking to pull out of a rough ending to 2018, when the company cut its outlook twice for the year’s final quarter because of weakness in China sales and in data-center and gaming sales. Part of the issue has been slow adoption of the company’s newest Turing chips, and Monday’s keynote was used to lay out more plans and adoption of the new equipment.

Nvidia also announced that more game companies were adopting real-time ray tracing, the most important advancement in the new Turing gaming chips, which was also tied to another conference happening an hour north — the annual Game Developers Conference in San Francisco. Nvidia said that Unreal Engine and Unity — two of the most popular gaming engines, which are important hubs for designers — had adopted real-time ray tracing.

Earlier in the month, Nvidia announced its largest acquisition ever, using nearly all the cash the company has on hand in a $6.9 billion bid for Mellanox Technologies Inc.














MLNX, -0.19%












 . Nvidia’s Huang said the move was to double down on the company’s bet on data centers, which makes up 31% of its sales, according to the company’s most recent quarter. That’s up from 21% of sales in the year-ago quarter.

Nvidia shares declined 0.5% to finish the regular session at $168.95, while the PHLX Semiconductor














SOX, -0.26%












 finished down 0.3% and the S&P 500 index














SPX, +0.37%












 closed up 0.3%. Shares were up about 0.2% in after-hours trading as the keynote was happening.

Get the top tech stories of the day delivered to your inbox. Subscribe to MarketWatch’s free Tech Daily newsletter. Sign up here.



Source link

LEAVE A REPLY

Please enter your comment!
Please enter your name here