IBM has rolled out a suite of performance management software to improve the responsiveness of underlying business intelligence applications.
The new DB2 Warehouse Performance Management Suite, an add-on to IBM's DB2 database management system, can help companies better manage underlying system resources in data warehouses to optimise performance, said Marc Andrews, program director of IBM Data Warehousing.
"As customers start to expand the use of their data warehouse environment and use BI for a broader set of applications and use cases across the organisation, they are starting to see an impact on the performance of the underlying data warehouses," Andrews said. "Now they are starting to try to deliver insight directly into the business activities that people are performing on a day-to-day basis. The data being put into data warehouses is growing rapidly and the number of applications that hit that data are growing. It makes it difficult for organisations to control their warehouse."
The performance management tools in the new suite focus on supporting the new DB2 Warehouse 9.5 extreme workload management features announced in October, Andrews added. The new performance tool analyses the underlying system resources used for BI applications and applies workload classes.
In addition to prioritising the types of BI applications and arranging them in a workload queue according to their importance, the software also makes sure that the appropriate resources will be in place to process the applications, Andrews said.
For example, while a call centre application needed to support a customer phone call will be prioritised before a batch processing of a report, the batch process may start running later and eat up system resources that may be needed for a subsequent call centre request. The software enables companies to manage the underlying resources to ensure that lower-priority applications won't take resources away from more critical needs, he added.
Andrews said that IBM integrated Appfluent Technology's monitoring and workload analysis software into the suite of tools to add performance monitoring capabilities. With the functionality from those tools, he said, companies can monitor which applications and users are accessing which data sets and how frequently they are using those data sets.
"If there is a set of data that hasn't been accessed in six months, you can push that off to be archived," Andrews said. Or a company can partition certain data sets that are needed in high volumes so they don't gobble up resources, he added.