The company, a subsidiary of NCR, announced Teradata Warehouse Version 7.0 on Monday. Among the product's 150 enhancements are improved performance and a common interface for ease of use.
The latest version will also allow end users launching complex searches of a data warehouse to access the application at the same time as users making more specific, smaller-scale queries, said Vickie Farrell, vice-president of Teradata warehouse marketing.
That will allow customers to collapse smaller databases, also known as data marts, into a single Teradata warehouse and cut down on the cost of maintaining and synchronising various business intelligence applications, she said.
The last version of the application allowed users to run a warehouse-wide, exhaustive query against historical data. Such "strategic" analytical operations might include an entire sales or marketing campaign.
Version 7.0 will let users make more limited queries that could, for instance, focus on one customer. It also allows existing data, along with long-term historical information, to be integrated into a query.
For example, a fraud-detection application at a credit card company might in real time detect a transaction by a user and check that purchase against all other records in the database to determine if it is aberrant and indicative of potential theft, said Farrell.
Other enhancements include a load-balancing feature that can prioritise time-sensitive queries without slowing down the overall performance of the application.
It can also partition chunks of related data - such as monthly or weekly sales records - and run queries on them on a single server processor, which speeds performance. Normally, the data has to be distributed across processors equally, which is more time-consuming, Farrell said.
Teradata Warehouse Version 7.0 will be available in December.