This summer marks the one-year anniversary of OpenStack, a cloud-focused open source effort initiated by Rackspace. Already OpenStack has seen 20,000 downloads, involvement from such major players as Cisco (News - Alert) and Citrix, and deployments by enterprises and top-level service providers including Internap, KT and NT.

Jonathan Bryce, chairman of the project policy board for OpenStack, says the effort evolved out of software Rackspace (News - Alert) used to run its private cloud services. Rackspace offers dedicated server, managed hosting and web hosting, and is a leader in cloud services, for which is has 100,000 customers. The company decided to open source the software in a move to increase adoption of cloud services as a whole, says Bryce, adding the bigger the pie the bigger the piece Rackspace can slice for itself.

However, just as Rackspace was readying to open its cloud source code, it heard that NASA had a similar effort under way with the Nebula cloud solution. So Rackspace joined forces with NASA under the umbrella of the cloud initiative now known as OpenStack.

Today about 60 companies are involved in OpenStack – either deploying or writing code based on it. And between 15 and 20 organizations, including Cisco and Citrix, have full-time developers dedicated to OpenStack. Bryce says OpenStack represents an “amazing community of technology experts” from different organizations around the world that are providing free software that can automate everything in a data center.

“That’s going to open up huge opportunities for innovation and to lower costs and to improve the state of data centers everywhere,” he adds.

Bryce says that while there were some cloud services out there before the launch of OpenStack, different players had different stacks and APIs, which was slowing down the adoption of cloud solutions because users didn’t give them the visibility and control they needed. He adds that customers also wanted to run cloud solutions inside their data centers on their networks, but a lot of available software wasn’t written at any kind of scale to allow for that. Also, some users wanted to leverage the public cloud for functions that weren’t required to be on site. The efforts related to OpenStack, he adds, address all of the above.

OpenStack on April 15 made available its third software release, called Cactus, which includes OpenStack Compute, Object Storage and the Image Service. Cactus is the first release in which all three of those components are production ready, Bryce says, adding that when OpenStack started it didn’t have an image service at all.

The OpenStack Compute module controls virtual servers and supports hypervisors from Citrix, Microsoft (News - Alert), VMware and the KVM open source hypervisor, among others. And it does it in an automated way so it can scale.

Object Storage is a distributed storage system meant for massive volumes of data. It’s meant to be used for things like user-generated content, archives and backup for large volume of data, but without the need to pay a premium for Fibre Channel storage. This software runs on a set of servers and replicates software to different servers, so users can enjoy the benefits of using low- cost, commodity hardware, but with affordable reliability.

OpenStack Image Service, meanwhile, ties together compute and object storage. It can take snapshots of running virtual machines and use that to do back up or to configure other servers the same way.

Next month OpenStack expects to introduce another release; this one will be called Diablo. It marks the move of OpenStack beyond just virual computing and into the area of networking. Specifically, vendors of networking gear such as routers and switches will, as part of the OpenStack Diablo effort, present APIs in their hardware to let users dynamically configure VLANs, set QoS and otherwise do traffic management.

Diablo will be discussed in more detail at the next OpenStack Design Summit scheduled for April 26 to 29 in Santa Clara, Calif.