Statistical techniques for online anomaly detection in data centers

Online anomaly detection is an important step in data center management, requiring light-weight techniques that provide sufficient accuracy for subsequent diagnosis and management actions. This paper presents statistical techniques based on the Tukey and Relative Entropy statistics, and applies them...

Full description

Saved in:
Bibliographic Details
Published in2011 IFIP/IEEE International Symposium on Integrated Network Management pp. 385 - 392
Main Authors Chengwei Wang, Viswanathan, K., Choudur, L., Talwar, V., Satterfield, W., Schwan, K.
Format Conference Proceeding
LanguageEnglish
Japanese
Published IEEE 01.05.2011
Subjects
Production
Online AccessGet full text

Cover

More Information
Summary:Online anomaly detection is an important step in data center management, requiring light-weight techniques that provide sufficient accuracy for subsequent diagnosis and management actions. This paper presents statistical techniques based on the Tukey and Relative Entropy statistics, and applies them to data collected from a production environment and to data captured from a testbed for multi-tier web applications running on server class machines. The proposed techniques are lightweight and improve over standard Gaussian assumptions in terms of performance.
ISBN:9781424492190
142449219X
ISSN:1573-0077
DOI:10.1109/INM.2011.5990537