Title: Binomial logit regression and centralised agent stochastic optimisation for privacy preserved load balancing in cloud

Authors: M. Jawahar; A. Sabari; S. Monika

Addresses: K S R Institute for Engineering and Technology, Tamil Nadu, India ' K.S. Rangasamy College of Technology, Tamil Nadu, India ' K.S. Rangasamy College of Technology, Tamil Nadu, India

Abstract: Load balancing and privacy preservation of data plays an important role in cloud. Few research works have been designed to perform load balancing on cloud server but, performance was not improved. In order to overcome such limitation, a binomial logit privacy preserved load balancing (BLPPLB) technique is proposed. At first, the request is sent from user to cloud server. BLPPLB Technique carried out binomial logit authentication based on user behaviour on cloud. Then, BLPPLB technique finds the intruder attacks and authorised users in cloud. During data accessing process, load balancing is performed through selecting optimal server among multiple servers for each user requests based on objective function using centralised agent based stochastic local search to provide the requested services. The experimental result shows that the BLPPLB technique is able to increase the load balancing efficiency with higher data confidentiality when compared to state-of-the-art works.

Keywords: binomial logit regression; centralised agent; cloud server; intruder attacks; load balancing; objective function; stochastic local search; user request; privacy reservation; user behaviour; authorised user; data accessing; data confidentiality.

DOI: 10.1504/IJISCM.2019.101647

International Journal of Information Systems and Change Management, 2019 Vol.11 No.1, pp.25 - 43

Received: 10 Apr 2018
Accepted: 08 Apr 2019
Published online: 19 Aug 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article