Teradata is hoping to gain ground in the hotly competitive data-warehousing and analytics market with QueryGrid, a new product that allows users to run a single SQL-based query across multiple data stores, from Teradata’s own system to Hadoop and rival databases.
QueryGrid will be available in the third quarter and is aligned with the release of Teradata Database 15, which was also announced Monday.
Instead of moving data, the queries are pushed into target systems and the processing occurs there, Teradata said. This gives users the ability to choose the best processing engine for a particular job while working within a single environment, according to Teradata.
The approach echoes SAP’s Smart Data Services for its HANA in-memory database and Business Warehouse, where remote stores are viewed as virtual tables.
QueryGrid’s features build on Teradata’s previously available Unified Data Architecture, which integrates Hadoop with Teradata’s data warehouse and Aster Discovery Platform.
Meanwhile, Teradata Database 15 is now available and features a wide range of improvements, including support for JSON (JavaScript Object Notation) data. JSON is widely used in machine-to-machine communications, a trend known commonly as the “Internet of things.” This support will allow customers to run large-scale analytics on machine-generated data.
Teradata 15 also introduces support for more scripting languages, including Perl, Ruby, Python and the statistical modeling language R.
Other new features include expanded support for analyzing three-dimensional geospatial data. For example, a developer could analyze the size and height of natural features and neighboring buildings to determine what their impact could be on views from various parts of a planned building.
In addition, Teradata 15 introduces a range of stability and workload management enhancements. More details are expected to be revealed Monday during Teradata’s Universe conference in Prague.