Automatic detection of performance anti-patterns in inter-component communications

Abstract

Performance problems such as high response times in software applications have a significant effect on the customer's satisfaction. In enterprise applications, performance problems are frequently manifested in inefficient or unnecessary communication patterns between software components originating from poor architectural design or implementation. Due to high manual effort, thorough performance analysis is often neglected, in practice. In order to overcome this problem, automated engineering approaches are required for the detection of performance problems. In this paper, we introduce several heuristics for measurement-based detection of well-known performance anti-patterns in inter-component communications. The detection heuristics comprise load and instrumentation descriptions for performance tests as well as corresponding detection rules. We integrate these heuristics with Dynamic Spotter, a framework for automatic detection of performance problems. We evaluate our heuristics on four evaluation scenarios based on an e-commerce benchmark (TPC-W) where the heuristics detect the expected communication performance anti-patterns and pinpoint their root causes.

Runtime metric meets developer - Building better cloud applications using feedback

“…Some parts of FDD, specifically the use cases more geared towards predicting performance problems before deployment, are based on prior work in the area of performance anti-patterns [41,46] and their automated detection. Our implementation of these predictive FDD use cases also leans heavily on prior work in static and dynamic program analysis [38].…”

AIM: Adaptable Instrumentation and Monitoring for Automated Software Performance Analysis

“…In previous work we utilized AIM to enable automatic, experiment-based detection of performance problems [7], [16]. In the following, we apply AIM in a new experiment-based scenario, using it for automation of measurements to derive resource demands for the calibration of an architectural performance model.…”

DynamicSpotter

“…DS has been applied in multiple case studies [8,9] showing promising results with respect to the automation of performance problems diagnostics. As DS is a framework, the diagnostics capabilities highly depend on the detection heuristics available for DS.…”

“…As many performance antipatterns share common characteristics, they can be structured in a hierarchical way, yielding a taxonomy which covers performance problem types from their high level symptoms to their specific root causes [8,9]. DS utilizes the taxonomy as a decision tree in order to systematically search for performance problems.…”