You might not have heard of Google’s Bigtable database, but you’re almost certainly a regular user. That’s because it’s the database that powers the majority of Google’s products, including search, Maps, Gmail, YouTube and more. It’s a huge and extremely powerful database capable of handling all kinds of data.

And now Google is putting it in the cloud. Its new Google Cloud Bigtable service is reported to be a considerably cheaper alternative to similar cloud-based NoSQL databases like Amazon Web Services’ DynamoDB. NoSQL (not only SQL) databases differ from relational databases in that they can handle all kinds of unstructured data, as well as regular old structured row-and-column data.

Google has sensibly made sure its Cloud Bigtable is able to handle vast amounts of information straight out the box by supporting the HBase API. HBase is a similar NoSQL database, essentially an open-source version of Bigtable that stores and serves up data in the Hadoop open-source file system, and is used by companies including Ebay and Pinterest.

Google’s Cloud Bigtable has been ramped up to give better performance, however.

“The write throughput per dollar on this product is three times what the standard HBase implementation would be,” Tom Kershaw, head of product management for storage, networking, and big data at the Google Cloud Platform, told VentureBeat.

Google clearly has big business customers with petabyte-sized datasets in mind for Cloud Bigtable. The company already offers Cloud Datastore on its App Engine platform, but according to Kershaw that’s Google’s “getting started NoSQL database”, which means Bigtable is for the big boys.

“[Cloud Bigtable] is designed for larger companies and enterprises where extensive data processing is required, and where workloads are more complex,” Cory O’Conner, a Google Cloud Platform product manager, told TechCrunch. “For example, if an organization needs to stream data into, run analytics on and serve data out of a single database at scale – Cloud Bigtable is the right system.”

It’s easy to see why that is, with Google promising that Bigtable can deliver single-digit millisecond latency and two times the performance per dollar compared to other NoSQL databases like HBase or Cassandra. And because the HBase API is supported, it can also be integrated with existing applications in the Hadoop ecosystem, as well as with Google’s Cloud Dataflow. Moreover, Google reckons it will take just seconds to set up a Bigtable cluster, with storage scaling itself automatically.

O’Connor told ZDnet that he expects the new service to have a big impact in areas like advertising, energy, financial services, telecoms and the Internet of Things.

The service is priced at 65 cents per node, the unit in which performance is provisioned. One node delivers up to 10,000 reads and writes per second, or 10MB/s of throughputs for scans where there are no individual reads and writes. Meanwhile storage on SSD costs 17 cents per gigabyte per month on a pay-as-you-go basis. Although not available yet, Google plans to offer hard disk storage at 2.6 cents per month in the near future.

Cloud Bigtable is now available in beta, open to all developers but without a service level agreement or technical support.

Photo Credit: Iain Browne via Compfight cc