With Cloud Bigtable, Google is launching a new range of NoSQL databases, as the name suggests, is powered by Bigtable data storage syste...
With Cloud Bigtable, Google is launching a new range of NoSQL databases, as the name suggests, is powered by Bigtable data storage system of the company, but with the addition of compatibility with the Apache HBase API - which in turn is based on Google's Bigtable project. BigTable is used by major services like Gmail, Google Search and Google Analytics, then what Google is launching is definitely an extensively tested service.
Google promises that Cloud Bigtable offer maximum latency to a single digit of milliseconds and 2 times the performance compared to competing systems of caliber HBase and Cassandra. Since the API supports Apache HBase, Cloud Bigtable can be integrated with any existing ecosystem Hadoop applications, but also supports Google Cloud Dataflow.
BigTable is a proprietary system based data compressed high-performance, developed by Google since 2004 and built on Google File System (GFS), on Chubby Lock Service, on SSTable and some other Google technologies. It is designed to be able to support loads in the petabyte level through hundreds or thousands of machines and to make it easy to add new machines to the system, using the new resources without any reconfiguration is necessary. The reasons that prompted Google to develop its database scalability and better control on the performance characteristics.
The creation of a cluster with Cloud Bigtable should only take a few seconds, and the scale space automatically according to user needs.
It is worth noting that this is not the first product of Google NoSQL cloud-based database. With Cloud Datastore, Google already offers a NoSQL datastore for developers on its platform App Engine. Also this service is based on Bigtable.
Cory O'Conner, a product of Google Cloud Platform manager, says that Cloud Datastore focuses on loading and reading of hard labor, and is ideal for web applications and mobile applications.
