News

The San Diego Supercomputer Center (SDSC) in partnership with IBM recently offered several days of on-site training in IBM's DB2 relational database software in both application development and database administration.

"To support the growing importance of databases in today's data-intensive scientific research at SDSC, we wanted to offer SDSC staff advanced training in database administration and application development on DB2," said Dave Archbell, director of SDSC's Advanced Database Projects Lab in the Data and Knowledge Systems (DAKS) program. "We're very pleased that IBM was able to partner with us in offering this important service."

The DB2 training was very well attended, with some 30 people showing up, almost twice the number originally anticipated. The sessions were taught by IBM instructor Patti Cartwright. "We were very fortunate to have her teaching," said Archbell. "In addition to the core curriculum, she's very experienced in database tuning and covered a lot of important dos and don'ts that will be extremely useful."

The training, which was designed to help experienced Java programmers take full advantage of DB2 and its capabilities, covered Programming Using Java (DB2 UDB for Linux, Unix, and Windows), including JDBC Programming I; JDBC Programming II; SQLJ Programming; Stored Procedures; Object-Relational Capabilities; Performance Considerations; and Visual Explain (graphical explain plan tool).

The database administration component, which was for DBAs with experience primarily in Single-Partitioned Databases, covered DB2 UDB Multi-Partitioned Environment for Single Partition DBAs, including Concepts; Installation; Creating Databases; Partitioning and Database Partition Groups; Data Placement; Creating Tablespaces; Creating Objects; Moving Data; Recovery (as in backup); Scaling; Performance and Application Considerations; and Monitoring and Problem Determination.

The DAKS Advanced Database Projects Lab provides data services to advance science. Research focuses on making data available to researchers via traditional methods and Application Program Interfaces that allow simple storage and retrieval of data, regardless of type, size, and physical location. The lab provides infrastructure for data mining, data warehousing, and query processing for projects such as the NSF project in "Correlating Heterogeneous Measurements" for the Cooperative Association for Internet Data Analysis, or CAIDA. -Paul Tooby