Welcome to the InfoSci Platform
Reference Hub4
Performance Degradation Detection of Virtual Machines Via Passive Measurement and Machine Learning

Performance Degradation Detection of Virtual Machines Via Passive Measurement and Machine Learning

Toshiaki Hayashi, Satoru Ohta
Copyright: © 2014 |Volume: 5 |Issue: 2 |Pages: 17
ISSN: 1947-9220|EISSN: 1947-9239|EISBN13: 9781466652767|DOI: 10.4018/ijaras.2014040103
Cite Article Cite Article

MLA

Hayashi, Toshiaki, and Satoru Ohta. "Performance Degradation Detection of Virtual Machines Via Passive Measurement and Machine Learning." IJARAS vol.5, no.2 2014: pp.40-56. http://doi.org/10.4018/ijaras.2014040103

APA

Hayashi, T. & Ohta, S. (2014). Performance Degradation Detection of Virtual Machines Via Passive Measurement and Machine Learning. International Journal of Adaptive, Resilient and Autonomic Systems (IJARAS), 5(2), 40-56. http://doi.org/10.4018/ijaras.2014040103

Chicago

Hayashi, Toshiaki, and Satoru Ohta. "Performance Degradation Detection of Virtual Machines Via Passive Measurement and Machine Learning," International Journal of Adaptive, Resilient and Autonomic Systems (IJARAS) 5, no.2: 40-56. http://doi.org/10.4018/ijaras.2014040103

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Virtualization is commonly used for efficient operation of servers in datacenters. The autonomic management of virtual machines enhances the advantages of virtualization. Therefore, for the development of such management, it is important to establish a method to accurately detect the performance degradation in virtual machines. This paper proposes a method that detects degradation via passive measurement of traffic exchanged by virtual machines. Using passive traffic measurement is advantageous because it is robust against heavy loads, non-intrusive to the managed machines, and independent of hardware/software platforms. From the measured traffic metrics, performance state is determined by a machine learning technique that algorithmically determines the complex relationships between traffic metrics and performance degradation from training data. The feasibility and effectiveness of the proposed method are confirmed experimentally.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.