Oracle Corp. announced today that its Globally Distributed Autonomous Database is generally available.

Aimed at customers that have strict data sovereignty requirements, usually because of regulations, the product has all the features of the Oracle Autonomous Database with additional controls over data distribution placement. Organizations can automatically distribute and store data anywhere in the world with the physical location being transparent to applications, Oracle said. Autonomous operation provides automatic scaling, self-repair, tuning, security, backups and updates.

The Distributed Autonomous Database is a single logical database management system horizontally partitioned or sharded. Each shard is an independent database instance that hosts a subset of the logical database’s data and can be placed in one or more regions on Oracle Cloud Infrastructure.

The DBMS uses a common catalog for automated shard deployment, centralized management of a sharded database and multishard queries using an automated query coordinator. A query or application request is routed directly to one or more shards using Oracle driver sharding keys.

Flexible data types

Oracle said the database supports nearly any data type, including JavaScript Object Notation, documents, media and vectors. It also supports relational, NoSQL and in-memory analytical data models. Existing SQL queries can run against the distributed data without modification.

“You’ve got all the data models, data unification capability and you can share the database and deploy it in as many countries as you need versus trying to operate silos, which becomes increasingly complicated,” said Steve Zivanic, global vice president of database and autonomous services product marketing.

Flexible sharding makes partitioning easier, improves query performance and enables more fine-grained compliance, said Wei Hu, senior vice president of development for high availability and emerging technologies. “Most distributed databases for one or two ways to shard data; we support six,” he said.  They include hash, range, list, two types of composite and custom shards.

“With the autonomous capabilities, we can do automatic scaling all the way down to look at database consumption at the individual shard level,” Zivanic said. “That’s a unique capability.”

Interest growing

Oracle executives said the company is seeing growing interest in distributed databases as global organizations contend with increasing local regulation.

“Instead of trying to manually manage threat detection and remediation, backups, replication and such across 20 or 30 databases, we’re essentially adding an autonomous AI-powered layer on top of that is much more streamlined,” Zivanic said.  “Any company operating in even just a few countries can set different policies to meet the requirements for specific countries and still manage it as a single database.”

He said Oracle uses tactics like distributed parallel querying and shard-level querying to improve performance across a globally distributed data set. The technology can also route queries to nodes with light workloads.

The Distributed Autonomous Database is available on Oracle Cloud Infrastructure and installed locally on Oracle Exadata Cloud@Customer. The distributed database can also be implemented on non-Oracle on-premises servers without autonomous features. The company also plans to make the product available in the Microsoft Corp. Azure cloud.

Photo: Flickr CC