Cloudscaling, the cloud infrastructure software shop, has released Open Cloud System (OCS) 2.0 for general availability in production deployments.

CTO Randy Bias says, "General availability of OCS version 2.0 is a major milestone. As far as we are aware, we're the only company to ship an elastic cloud infrastructure software solution built on OpenStack technology for public or private deployments."

The new version adds OpenStack Folsom support and a new scale-out elastic block storage option.

OCS Block Storage is an Amazon-like network-attached disk solution that provides persistent, expandable and elastic block-level storage volumes. Volumes can be attached to a running instance and exposed as a device in the instance.

Cloudscaling has also developed an OCS Block Storage scheduler that maximizes volume dispersion on a per-tenant basis, improving fault tolerance by reducing the impact of infrastructure failures on tenant volumes.

Cloudscaling customers include enterprises, web application providers and service providers such as KT and Internap. The company said that sales activity has accelerated since version 2.0 was unveiled in October and to pick up some of the slack the company hired Andy Green as VP, business and channel development, a role he previously filled at Wyse, The Santa Cruz Operation and Sun Microsystems.

The company is a charter gold corporate sponsor of the OpenStack Foundation, and Bias has served on the board since its formation. Cloudscaling is a top-10 code contributor, including ZeroMQ messaging, RPC abstraction layer, APIs for Google Compute Engine and security improvements.

Cloudscaling says a major improvement in Folsom is the Nova compute's VM state management. Prior to Folsom, it was common to find a VM permanently stuck in a pending state with no way to recover using Nova API calls. The only way to get rid of a VM in this state was to manually update Nova's MySQL database to wipe out any trace of that VM, a potentially tricky and risky maneuver. In Folsom, it's less likely for VMs to become stuck and if they do, there's a "reset-state" API call that can be used to safely reset the VM state.