OpenStack Delivers First Release of Cloud Computing Tech

By Sean Michael Kerner (Send Email)
Posted Oct 21, 2010


More on cloud computing

After only three short months of life, the OpenStack open source cloud computing initiative is out with its first public release of production quality code. An effort originally started by NASA and Rackspace three months ago is now releasing production-ready code as outside contributions pour in.

The first OpenStack release is codenamed Austin and includes both storage and cloud compute fabric technologies that can be used by enterprises to deliver cloud services. Originally an effort kick-started by NASA and Rackspace, OpenStack is now benefiting from the support and contributions of more than 35 technology vendors.

"We now have our first software release -- where we saw a lot of interest initially in getting people involved. This is the first output of that effort," Mark Collier, vice president of community and business development, at OpenStack told InternetNews.com. "So for us it's about going from the promise of OpenStack, to the first chance for enterprises and service providers to kick the tires on the code and start planning for public and private cloud deployments."

The OpenStack project was first announced in July as an effort to provide new open source based cloud computing technologies.

OpenStack is currently comprised of an object store technology for cloud-based storage and a compute technology for cloud deployments. On the compute side, the Austin release of OpenStack will include support for the Xen virtualization hypervisor, thanks to code contributions from Citrix. The initial code drop of the compute engine from NASA had support for only the KVM hypervisor. Collier noted that the goal of OpenStack is to support multiple hypervisors, although currently the release build is lacking support for both VMware and Microsoft's Hyper-V.

"One of our core design philosophies is to make sure that whenever possible, people deploying OpenStack would have their choice of components," Collier said. "The first big area was to make it abstracted, so we could support multiple hypervisors. We do expect forthcoming support for other hypervisors, none that we can announce today."

On the compute side, there is also a REST-based OpenStack API intended to help enable developers to access the underlying platform features.

From a cloud storage perspective, the OpenStack Austin release builds on capabilities that Rackspace has already been deploying as part of its CloudFiles storage services. With OpenStack Austin, Collier noted some additional access controls have been added. The OpenStack project has also added user defined meta-data, so users can make their files and folders easier to identify.

Moving forward, one of the key challenges facing the OpenStack project as it continues to grow and mature revolves around how to actually handle the large amount of interest the project has generated. Collier noted that the interest has been overwhelming, and many developers and organizations have made offers of code contributions.

"For us it's a question of discipline and making sure we don't get ahead of ourselves," Collier said. "We have to take a methodical approach to adding features that are necessary."

Collier added that one of those necessary features likely to debut in a future OpenStack release is some kind of live migration support for cloud services and content.

"If you think about the promise of having lots of clouds that are compatible, you want to be able to move things between them," Collier said.

Sean Michael Kerner is a senior editor at InternetNews.com, the news service of Internet.com, the network for technology professionals.Follow ServerWatch on Twitter

Page 1 of 1


Comment and Contribute

Your name/nickname

Your email

(Maximum characters: 1200). You have characters left.