Stefanie Scherzinger's talk on DB Seminar

Tales from Industry: Data Stream Processing in Database Performance Monitoring

 
Abstract
This talk features the IBM Infosphere Optim Performance Manager (OPM), a commercial tool that is capable of monitoring the performance of DB2 instances in distributed environments. OPM closely tracks the execution of database workload, from the point when the workload is issued in a database application, to its actual execution on the database server. This allows to break down the response time as experienced by the user into time spent waiting for connections, within database drivers, the network, and finally, within the database server. From a data management point-of-view, the core challenge is the handling of large amounts of performance data, arriving as streams, and the implementation of window-based joins to match monitoring data from different sources. In this talk, I introduce the architecture of OPM, the data flow in collecting end-to-end monitoring data, and share my experiences as a software engineer and the feature owner for the data stream processing on the OPM development team.
 
Bio
Stefanie Scherzinger recently joined the Regensburg University of Applied Sciences as a new faculty member. Her professional and teaching interests are in databases and programming languages. Prior to joining academia, Stefanie Scherzinger engineered software for IBM and Google, where she could not publish but instead patented her ideas. She holds a PhD from Saarland University, contributing to the efficiency of main-memory-based XML query engines.