ABSTRACT
In performance tuning of computing systems, identifying bottlenecks which limit the performance of the system is essential. A variety of factors can cause the bottleneck, and therefore, identification and removal of the bottleneck becomes complicated. This paper proposes a technique which supports identifying the bottleneck together with its cause. In the proposed technique, each component in the target system is automatically tuned by adjusting configuration parameters. The effect and the applicability of the parameter tuning are locally analyzed at each component to diagnose whether the bottleneck exists in the tuned component or not. The diagnosis results are collected from each component in the target system, and then, the bottleneck in the whole system can be identified. The bottleneck caused by resource shortages can be identified based on the applicability of the parameter tuning. The effect of the parameter tuning is useful to identify the bottleneck caused by the other external factors. Therefore, the cause of the bottleneck can be classified as either resource shortages or the other factors.
- Y. Diao, J. L. Hellerstein, S. Parekh, and J. P. Bigus. Managing web server performance with autotune agents. IBM Syst. J., 42(1): 136--149, 2003. Google ScholarDigital Library
- T. Osogami and S. Kato. Optimizing system configurations quickly by guessing at the performance. In Proceedings of the 2007 ACM SIGMETRICS international conference on Measurement and modeling of computer systems (SIGMETRICS'07), pages 145--156, 2007. Google ScholarDigital Library
Index Terms
- Identifying performance bottlenecks based on the local parameter tuning
Recommendations
Fast, automatic, procedure-level performance tuning
PACT '06: Proceedings of the 15th international conference on Parallel architectures and compilation techniquesThis paper presents an automated performance tuning solution, which partitions a program into a number of tuning sections and finds the best combination of compiler options for each section. Our solution builds on prior work on feedback-driven ...
When parameter tuning actually is parameter control
GECCO '11: Proceedings of the 13th annual conference on Genetic and evolutionary computationIn this paper, we show that sequential parameter optimization (SPO), a method that was designed for (offline) parameter tuning, can be successfully used as a controller for multistart approaches of evolutionary algorithms (EA). We demonstrate this by ...
Context-sensitive delta inference for identifying workload-dependent performance bottlenecks
ISSTA 2013: Proceedings of the 2013 International Symposium on Software Testing and AnalysisSoftware hangs can be caused by expensive operations in responsive actions (such as time-consuming operations in UI threads). Some of the expensive operations depend on the input workloads, referred to as workload-dependent performance bottlenecks (...
Comments