van Hoorn, Andre (2015) Measurement-based Performance Problem Detection and Diagnosis. [Invited talk] In: Invited talk at Gran Sasso Science Institute. , April 21, 2015, L'Aquila, Italy .

Preview

Slideshow 20150421-GSSI-LAquila.pdf Download (4MB) | Preview

Abstract

Application performance monitoring (APM) is getting more and more common in practice. The APM data obtained from the monitored application systems – ranging from aggregated response time and resource utilization values to detailed event traces of individual transactions – includes valuable information for detecting and diagnosing performance problems even before they occur. However, particularly problem diagnosis is currently often performed manually with little guidance by APM tools. After a brief introduction into application performance problem detection and diagnosis, I will present what we have done in this area in the context of our open-source APM framework Kieker (http://kieker-monitoring.net/) and outline our plans for the recently started collaborative research project diagnoseIT on expert-guided automatic diagnosis of performance problems in enterprise applications.

Actions (login required)

View Item