OpenStack now does NoSQL

Chalk up another notch for OpenStack’s growth as a platform. Tesora, an OpenStack developer outfit that focuses on the Trove database component, is announcing partnerships with both Red Hat and MongoDB, the better to give OpenStack the power to manage NoSQL solutions.

OpenStack’s Trove subcomponent was created to allow databases to be managed within OpenStack with the same flexibility and transparency as any other resource, such as storage or networking. It’s not an experimental technology, either; it’s being used right now by eBay and Rackspace in live production.

Until now, the only kind of database that could be provisioned under OpenStack was MySQL. Starting there made sense, sinceMySQL is ubiquitous, well-understood, and the most likely candidate for the kind of widescale management and provisioning performed through OpenStack. But it’s also only one database among many — PostgreSQL, for instance, is regarded as being far more robust than MySQL. And while PostgreSQL will eventually be manageable under OpenStack, there’s no timetable yet for when such management will arrive.

Tesora’s plan is to have MongoDB be manageable under Trove as a first-class citizen, allowing instances of both databases to run side by side within OpenStack. As with MySQL, MongoDB is hardly the only NoSQL solution of its kind out there; CouchDB, Riak, and Couchbase all come to mind. But it’s next to impossible to go wrong by starting there, given MongoDB’s broad adoption as compared to its competitors, not to mention the burgeoning demand for MongoDB expertise.

The competition here for OpenStack (and its vendors) comes most directly in the form of NoSQL-as-a-service via outfits like Rackspace. Despite all that its vendors have done to make OpenStack easier to wrap one’s head around and deploy, OpenStack still suffers a bit from the perception that it’s underadopted and puts an exceptional burden on the folks deploying it. If an enterprise can get enough of the same functionality, granularity, and power from a third-party service provider as it can from OpenStack — and with less sweat and muscle expended — then all the database support in the world for it won’t help much.

Another indirect source of competition, at least as far as the delivery of managed NoSQL goes, may come from IBM. Big Blue’s shifted its own cloud empire to be built on OpenStack, but it isn’t shirking any easy chances to monetize that process. Its recent acquisition of Cloudant lets IBM provide NoSQL-as-a-service in the form of CouchDB, another major name-brand NoSQL database. This isn’t likely to lure away anyone already committed to MongoDB in whatever manner, but it shows how yet another set of options exist, both in terms of where to get it and in what form.

For those looking for more hands-on information, Tesora has a panel planned at the upcoming OpenStack Summit in Atlanta, where it intends to show how both MySQL and MongoDB can be deployed and managed via Trove.

This story, “OpenStack now does NoSQL,” was originally published at InfoWorld.com. Get the first word on what the important tech news really means with the InfoWorld Tech Watch blog. For the latest developments in business technology news, follow InfoWorld.com on Twitter.