Intel and Google Cloud to develop new class of data center chip

Intel and Alphabet’s Google Cloud have worked together to create a new category of chip that Intel hopes will become a major seller in the booming cloud computing market.
Google Cloud at IT trade show
The new chip called Mount Evans and will be sold to others beyond Google, reflects the way that cloud computing providers operate. They build huge data centers full of powerful physical computers and sell virtual slices of those machines to other businesses, who in turn get better bang for the buck than building the machines themselves.

For cloud providers, tasks like setting up the virtual machines and getting customer data to the right place are essentially overhead costs. The Mount Evans chip, which Google and Intel have dubbed an “infrastructure processing unit” (IPU), separates those tasks out from the main computing tasks and speeds them up. Doing so helps ensure the safety of those functions against hackers and adds flexibility to the data center.

“We see this as strategically vital. It’s an extremely important area for us and for the data center,” Nick McKeown, senior vice president of the network and edge group at Intel, told Reuters.

Intel is not the only player making infrastructure chips. Nvidia and Marvell Technology have similar but slightly different offerings.

But Intel and Google are working together on a set of software tools that will be released for free in hopes of making Intel’s version of the chip a broader industry standard used beyond Google’s data centers.

Google is hoping to spur a technology trend that makes it easier for all data center operators to be more flexible about how they slice up their physical computer servers into virtual ones to suit whatever computing task is at hand, Amin Vahdat, a Google fellow and technical lead for networking at the company, said.