LinkedIn, which Microsoft has decided to acquire for more than $26 billion, today announced the formation of Open19, a project in which LinkedIn will work with partners to develop a new standard for data center infrastructure, specifically servers.
“The goals of Open19 are to provide lower cost per rack, lower cost per server, optimized power utilization, and (eventually) an open standard that everyone can contribute to and participate in,” Yuval Bachar, principal engineer for global infrastructure architecture and strategy at LinkedIn, and formerly a hardware engineer at Facebook, wrote in a blog post.
Given that description, the project bears some resemblance to the Open Compute Project (OCP), an initiative that Facebook established in 2011 and that has since received participation from Apple, Microsoft, Google, and several vendors.
Open19 — the name refers to the goal of fitting into any 19-inch server rack — has gotten buy-in from Hewlett-Packard Enterprise (HPE), Hyve, Mellanox, and Wiwynn, among others. All four of those companies have done work with the OCP.
There’s no question that LinkedIn knows what it’s doing in operating data center infrastructure. The core app has more than 433 million users, and it’s fair to say LinkedIn operates at web scale right alongside Facebook, Google, and Twitter. Surely, some companies will be keen to try out the style of server that LinkedIn has come up with. (LinkedIn intends to use the Open19 server design in its own data centers, Bachar wrote in today’s blog post.) But it is worth wondering why LinkedIn chose to set up a new organization instead of just joining forces with the OCP.
“My understanding is we just wanted to make sure it’s in a form factor that would be more easily adopted by more people,” a LinkedIn spokesperson told VentureBeat.
Facebook did not immediately respond to a request for comment.