Intel, Rackspace Team To Drive OpenStack Innovation

Intel and Rackspace unveiled a partnership Thursday aimed at making it easier to deploy, manage and use OpenStack in the hope of advancing enterprise adoption of the open-source cloud operating system.

The project entails creation of the OpenStack Innovation Center, a new R&D facility to be housed in Rackspace's headquarters in San Antonio, where the world's largest assembly of OpenStack developers will come together.

Rackspace and Intel also will deploy a test computing cluster of 2,000 server nodes that will be available to OpenStack developers from outside the project.

[Related: The Great OpenStack Debate: A Powerful Platform That Will Revolutionize The Data Center? Or A Dud?]

The pool of engineers will focus on maturing the open-source code, making it easier to use and more enterprise-ready, said Scott Crenshaw, senior vice president of strategy and product at Rackspace.

What they won't be doing is creating a new distribution.

"It's not a fork," Crenshaw said. "One of the things that brings Intel and Rackspace together are these are the two largest companies that are not trying to make an OpenStack distribution."

The project ultimately will make it easier for solution providers and their customers to consume OpenStack, he said.

Advances will put OpenStack on a path "to unlock the power and the value-add of the VAR channel," Crenshaw told CRN. "Our hope is the channel, over time, will become a significant part of the OpenStack value chain. I think it's been held back in the past."

OpenStack is on a roll, growing by 40 percent this year, Crenshaw told CRN, faster than the IaaS market as a whole. The collaboration, he said, will spur thousands of new cloud deployments in hybrid environments.

Crenshaw told CRN that hundreds of engineers -- some from Intel, others that will be recruited -- while housed at Rackspace headquarters will be working upstream on the project, contributing to open-source code for the entire OpenStack community.

Those bound for San Antonio without much OpenStack experience will learn faster "by sitting next to Rackspace experts," Crenshaw said.

Those engineers will be building enterprise features, fixing bugs and focusing on high-availability and disaster recovery.

OpenStack is notoriously difficult to work with, especially at large scales.

As the operator of the largest OpenStack cloud in the world and one of the project's original founders, however, Rackspace knows how to support and scale the technology, which it does for private cloud customers as well.

"It's hard for ordinary companies to recruit and retain and train people on specialized platforms," he said.

Darrin Hansen, general manager of Rackspace's private cloud business, said the 2,000-node test cluster will be available to "anyone testing, experimenting with new code who wants to scale it and see it run in a larger environment," not just Intel and Rackspace developers.

"We want the community to be with us on this. For the entire OpenStack family to benefit from this," Hansen told CRN.

The facility will have a front end that allows reservations and scheduling resources for running test code, he said.

In a phone conference announcing the partnership, Jason Waxman, general manager of Intel's Cloud Infrastructure Group, said further maturation of OpenStack will spawn a diverse number of cloud deployments.

"Most people look at cloud and think the market is on fire. We're in a position to say, 'You know what, it's not moving fast enough,' " Waxman said.

Intel, Santa Ana, Calif., wanted to partner with Rackspace because it was a founder of OpenStack, and the company has the most experience running it at scale.

Diane Bryant, general manager of Intel's Data Center Group, said the OpenStack ecosystem needs development across the software-defined infrastructure stack to deliver a product that's functional and easy to deploy.

"Success is implementation of a full-service cloud portal in one day," Bryant said, adding that deployment should really be possible in an hour.

The center, over the next six months, will produce features such as a better scheduler, improved networking, new container services, and general bug fixes to make sure the code as a whole is more robust.

"Our goal is to enable enterprise-class features at scale of thousands of nodes," Bryant said. "We want to make sure the overall project is healthy."

Vendors of OpenStack distributions such as Red Hat, Mirantis, Hewlett-Packard and Huawei will benefit from the work that will be done in San Antonio, she added.

"We at Intel feel compelled that we need to take this action," Bryant said.

PUBLISHED JULY 24, 2015