Self-adaptive software system monitoring for performance anomaly localization

Abstract

Autonomic computing components and services require continuous monitoring capabilities for collecting and analyzing data of runtime behavior.Particularly for software systems, a trade-off between monitoring coverage and performance overhead is necessary.In this paper, we propose an approach for localizing performance anomalies in software systems employing self-adaptive monitoring. Time series analysis of operation response times, incorporating architectural information about the diagnosed software system, is employed for anomaly localization. Comprising quality of service data, such as response times, resource utilization, and anomaly scores, OCL-based monitoring rules specify the adaptive monitoring coverage. This enables to zoom into a system's or component's internal realization in order to locate root causes of software failures and to prevent failures by early fault determination and correction.The approach has been implemented as part of the Kieker monitoring and analysis framework. The evaluation presented in this paper focuses on monitoring overhead, response time forecasts, and the anomaly detection process.

Microservices for Scalability

“…Live visualization with ExplorViz is scalable [6] and elastic in cloud environments [28]. Monitoring may provide runtime models [23] for system comprehension [9], trace visualization [4], architecture conformance checks [11], and a landscape control center [12] with performance anomaly detection [3,24]. New perspectives on employing virtual reality [8] and physical models [7] are further explored.…”

AIM: Adaptable Instrumentation and Monitoring for Automated Software Performance Analysis

“…However, this cannot be assumed in many cases. Based on Kieker, Ehlers et al [5] propose an approach for self-adaptive monitoring to localize performance anomalies. Following a similar idea as described in [6], the authors fully instrument the AUT with switchable measurement probes.…”

“…However, they lack support for adaptation of the instrumentation, hence rendering automation of measurement-based performance engineering tasks impractical. Existing approaches which provide means for adaptation of instrumentation are limited in their adaptation capabilities [5], [6] as they are based on a switch-based principle, allowing to switch between instrumentation instructions injected into the AUT at startup. However, this is a rather inflexible approach limiting the adaptation of instrumentation instructions.…”

Including Performance Benchmarks into Continuous Integration to Enable DevOps

“…The actual detection of the anomalies has to be performed manually. However, previous work on anomaly detection within Kieker results can be adapted for this scenario [1].…”

“…The final three builds demonstrate our bug fixing two and a half month after the performance regression. With the help of our presented vision of including benchmarks into continuous integration and performing automated anomaly detections on the results, e. g., similar to [1], the time to fix performance regressions can be reduced.…”