Institutional Repository
| Detecting performance anomaly with correlation analysis for Internetware | |
| Wang, Tao (1); Wei, Jun (1); Qin, Feng (4); Zhang, WenBo (2); Zhong, Hua (2); Huang, Tao (1); Wei, J.(wj@otcaix.iscas.ac.cn) | |
| 2013 | |
| Source | Science China Information Sciences
 |
| ISSN | 1674733X |
| Volume | 56Issue:8Pages:1-15 |
| English Abstract | Internetware has become an emerging software paradigm to provide Internet services. The performance anomaly of Internetware services not only affects user experience, but also causes severe economic loss to service providers. Diagnosing performance anomalies has become one of the keys to improving the quality of service (QoS) of Internetware. Existing approaches create a system model to predict performance. Then, the prediction from the model is compared with the observation; a significant deviation may signal the occurrence of a performance anomaly. However, these approaches require domain knowledge and parameterization efforts. Moreover, dynamic workloads affect the accuracy of performance prediction. To address these issues, we propose a correlation analysis based approach to detecting the performance anomaly for Internetware. We use kernel canonical correlation analysis (KCCA) to model the correlation between workloads and performance based on monitoring data. Furthermore, we detect anomalous correlation coefficients by XmR control charts, which detect the anomalous coefficient and trend without a priori knowledge. Finally, we adopt a feature selection method (Relief) to locate the anomalous metrics. We evaluated our approach on a testbed running the TPC-W industry-standard benchmark. The experimental results show that our approach is able to capture the performance anomaly, and locate the metrics relating to the cause of anomaly. © 2013 Science China Press and Springer-Verlag Berlin Heidelberg.; Internetware has become an emerging software paradigm to provide Internet services. The performance anomaly of Internetware services not only affects user experience, but also causes severe economic loss to service providers. Diagnosing performance anomalies has become one of the keys to improving the quality of service (QoS) of Internetware. Existing approaches create a system model to predict performance. Then, the prediction from the model is compared with the observation; a significant deviation may signal the occurrence of a performance anomaly. However, these approaches require domain knowledge and parameterization efforts. Moreover, dynamic workloads affect the accuracy of performance prediction. To address these issues, we propose a correlation analysis based approach to detecting the performance anomaly for Internetware. We use kernel canonical correlation analysis (KCCA) to model the correlation between workloads and performance based on monitoring data. Furthermore, we detect anomalous correlation coefficients by XmR control charts, which detect the anomalous coefficient and trend without a priori knowledge. Finally, we adopt a feature selection method (Relief) to locate the anomalous metrics. We evaluated our approach on a testbed running the TPC-W industry-standard benchmark. The experimental results show that our approach is able to capture the performance anomaly, and locate the metrics relating to the cause of anomaly. © 2013 Science China Press and Springer-Verlag Berlin Heidelberg. |
| Indexed Type | SCI ; EI |
| Keyword | Performance Anomaly Anomaly Detection Internetware System Metrics Kernel Canonical Correlation Analysis |
| Department | (1) State Key Laboratory of Computer Science, Beijing, 100190, China; (2) Institute of Software, Chinese Academy of Sciences, Beijing, 100190, China; (3) University of Chinese Academy of Sciences, Beijing, 100049, China; (4) The Ohio State University, Columbus, OH, 43210, United States |
| Language | 英语 |
| WOS ID | WOS:000323665900005 |
| Citation statistics | |
| Content Type | 期刊论文 |
| URI | http://ir.iscas.ac.cn/handle/311060/16920 |
| Collection | 中国科学院软件研究所 |
| Corresponding Author | Wei, J.(wj@otcaix.iscas.ac.cn) |
| Recommended Citation GB/T 7714 | Wang, Tao ,Wei, Jun ,Qin, Feng ,et al. Detecting performance anomaly with correlation analysis for Internetware[J]. Science China Information Sciences,2013,56(8):1-15. |
| APA | Wang, Tao .,Wei, Jun .,Qin, Feng .,Zhang, WenBo .,Zhong, Hua .,...&Wei, J..(2013).Detecting performance anomaly with correlation analysis for Internetware.Science China Information Sciences,56(8),1-15. |
| MLA | Wang, Tao ,et al."Detecting performance anomaly with correlation analysis for Internetware".Science China Information Sciences 56.8(2013):1-15. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment