Welcome to the InfoSci Platform
Reference Hub1
Predicting Software Aging With a Hybrid Weight-Based Method

Predicting Software Aging With a Hybrid Weight-Based Method

Yongquan Yan, Yanjun Li, Bin Cheng
Copyright: © 2021 |Volume: 14 |Issue: 4 |Pages: 12
ISSN: 1938-7857|EISSN: 1938-7865|EISBN13: 9781799860037|DOI: 10.4018/JITR.2021100105
Cite Article Cite Article

MLA

Yan, Yongquan, et al. "Predicting Software Aging With a Hybrid Weight-Based Method." JITR vol.14, no.4 2021: pp.58-69. http://doi.org/10.4018/JITR.2021100105

APA

Yan, Y., Li, Y., & Cheng, B. (2021). Predicting Software Aging With a Hybrid Weight-Based Method. Journal of Information Technology Research (JITR), 14(4), 58-69. http://doi.org/10.4018/JITR.2021100105

Chicago

Yan, Yongquan, Yanjun Li, and Bin Cheng. "Predicting Software Aging With a Hybrid Weight-Based Method," Journal of Information Technology Research (JITR) 14, no.4: 58-69. http://doi.org/10.4018/JITR.2021100105

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Since software aging problems have been found in many areas, how to find an optimal time to rejuvenate is vital for software aging problems. In this paper, the authors propose a newly hybrid method to predict resource depletion of a web server suffered from software aging problems. The proposed method comprises three parts. First, a smoothing method, self-organized map, is used to make resource consumption series glossier. Second, several sub-optimal methods are utilized to fit resource consumption series. Third, an optimization method is proposed to combine all single methods to predict software aging. In experiments, the authors use the real commercial running dataset to validate the effect of the proposed method. And the presented method has a better prediction result for both available memory and heap memory under two metrics: root mean square error and mean average error.

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.