IBM Tuesday put out four tools to help with the programming, data mining and replication monitoring of its DB2 Universal Database, company officials announced.
The data-mining tool is WebSphere Studio Plug-in for DB2 Intelligent Miner, a graphical front-end for DB2 Intelligent Miner Scoring/Modeling. It allows users to add data-mining abilities to applications. Officials said in a statement that the plug-ins interface is intuitive enough to use without developers having to learn Intelligent Miners SQL API.
The plug-in features a view of data-mining models stored in the database. This view allows for import, export, renaming, deletion and visualization of data and enables a user to draw the flow of a data-mining process.
IBMs new replication-monitoring tool is called Replication Monitoring Center for DB2 (RMCDB2). Its geared for supervision of replication and feeds reporting configuration and operational data back to system operators in structured views. RMCDB2 supports alert monitoring that reports problems via e-mail, customized user exit routines or an application alert window.
The tool can be used to perform housekeeping tasks such as triggering a full refresh for data replication; changing replication activation and timing; adding or removing replication events; changing operational parameters for the run-time compo-nents Capture and Apply and the Replication Alert Monitor; and pruning component traces and histories. RMCDB2 works with DataPropagator Versions 5, 6, 7 and 8.
To let users get at the database more easily, IBM released a trial version of URI Access for DB2, a servlet that uses various access methods to address and retrieve data and metadata stored in DB2. To use the servlet, users provide URIs (Uniform Resource Identifiers) for access to DB2 instances through Web browsers and other clients that use the supported protocols, such as XML parsers. It now only supports HTTP.
Finally, IBM, of Somers, N.Y., offers a SQL interface for the main steps of data mining. The Easy Mining Procedures for DB2 Intelligent Miner are implemented as Java stored procedures in DB2. According to IBMs alphaWorks site, the procedures work with Intelligent Miner Modeling and Scoring, with one call to an Easy Mining procedure corresponding to the execution of several SQL statements that rely on the functions and methods of Intelligent Miner Modeling and Scoring.
The Easy Mining Procedures for DB2 Intelligent Miner contain stored procedures for basic mining functions, such as classification clustering or discovery of association rules. It covers more higher-level mining functions as well, including prediction-model discovery. Trial versions of the tools are available for free through IBMs alphaWorks site.