Teradata for the first time has separated its market-leading data warehouse software from its hardware and is offering it in two free Express versions aimed at development projects and cloud computing.

Teradata Express on Amazon EC2 is a version available for developers to self-provision on Amazon's Elastic Compute Cloud. The cost to customers paying Amazon's hourly fee for server use, is about $.10 an hour.

The use of the Teradata software on EC2 is free and good for working with up to one terabyte of data. Teradata runs under Novell Suse Linux Enterprise Server 10. Use of Teradata in this setting is designed for non-production evaluation workloads and for developers working on data warehouse projects, said Dan Graham, data warehouse program director, in an interview. The availability of Teradata in the cloud for modeling data warehouses and building applications appears to be a response to the growing competition in the data warehouse machine market. Teradata has been the leader the market leader for many years, but Oracle, IBM and Netezza are seeking to compete there as well.

Cloud computing could also one day offset database machines. Teradata appears to no longer wish to be seen as a strictly database machine company.

The other freely available version of Teradata is Teradata Express for VMware Player. VMware Player is a free desktop runtime environment for virtual machines from VMware. Teradata Express can be loaded into it under SUSE Linux and used for testing data warehouse systems running in an enterprise cloud on x86 servers.

"VMware is an important stepping stone towards delivering internal private clouds for developers and information technology operations managers," said Graham. With Teradata Express, they can build a demonstration data warehouse on a cluster of x86 servers, he said. Teradata has added a feature to its regular data warehouse machine product line as well, the Teradata 13 Database. The addition of Teradata Agile Analytics Cloud is intended to give data warehouse managers the ability to rapidly prototype an analytics datamart inside the larger warehouse. The mart occupies a sandbox where it can be tested and explored without violating other data warehouse operations, Graham said.

Agile Analytics makes use of a new Teradata Elastic Mart Builder tool to construct a data mart, and test or use it with real data already resident in the data warehouse, while keeping the mart's operations separate from the production system. The marts will have size limits and expiration dates so they don't proliferate and choke the data warehouse.

The new feature is a free addition to the main system and is meant to give data warehouse administrators a rapid prototyping system to satisfy end users without adding another Teradata system or watching business units construct their own impromptu data marts. "We're not trying to accumulate datamarts with this new feature. We're trying to eliminate the shadow IT department that's building datamarts all over the organization," said Graham.

InformationWeek Analytics has published a report on the 10 steps to effective data classification. Download the report here (registration required).