Intel Bets on AI for Data Centers with Nervana Purchase
Intel is buying artificial intelligence startup Nervana Systems, placing a bet that AI technology and machine learning will underpin the next wave of corporate computing and give it a leg up in the data center market.
According to technology news site Recode, the deal is worth $408 million, a large sum for a two-year-old startup. Diane Bryant, executive vice president and general manager of the Data Center Group at Intel, wrote in a company blog post that Nervana "has a fully-optimized software and hardware stack for deep learning. Their IP and expertise in accelerating deep learning algorithms will expand Intel’s capabilities in the field of AI."
Jason Waxman, vice president and general manager of the Cloud Platforms Group at Intel, told Recode that that the move to artificial intelligence could be a bigger shift in corporate computing than the cloud. Intel sees deep learning technology, which involves the use of algorithms to learn to recognize patterns, as necessary in a world of billions of connected devices generating massive amount of data.
“There is far more data being given off by these machines than people can possibly sift through,” Waxman told Recode.
As Recode notes, Nervana’s approach to AI and deep learning likely appealed to Intel since it does not just focus on software but also been working to infuse silicon chipsets with machine learning capabilities.
Although Intel largely missed out providing chipsets for smartphones, it has a strong position in the data center market, and the company thinks that AI inside data centers is the next big thing. Indeed, Google is using machine learning AI technology it acquired from DeepMind Technologies in 2014 to boost data center energy efficiency.
“There's always a next wave,” Waxman told Recode, adding that that corporate computing has already gone from mainframes to client-server and now on to cloud computing. “I firmly believe this is not only the next wave but something that will dwarf the last wave.”
