Title: Density-aware replica server placement for utilisation enhancement

Authors: Darothi Sarkar; Nitin Rakesh; Krishna Kumar Mishra; Anupama Mehra

Addresses: Institute of Engineering and Management, Kolkata, India; University of Engineering and Management, Kolkata, India ' Department of Computer Science and Engineering, Sharda University, Uttar Pradesh, India ' Department of Computer Science and Engineering, Motilal Nehru National Institute of Technology, Allahabad, India ' Department of Electronics and Communication Engineering, Amity University, Uttar Pradesh, India

Abstract: The most challenging issue in content delivery networks (CDN) is to place the surrogate servers. K-means clustering can be chosen as a very simple and effective approach for placing the surrogates over the network. But the main concern of this approach is that it may produce clusters with very few nodes or no nodes. In CDN, this scenario will lead to a situation where a surrogate is serving no clients. This paper introduces a parameter called population threshold for each cluster and also optimises the number of surrogates by restricting the placing of servers in those under populated clusters. This work also identifies nearby servers which can serve the requests from clusters with population below the defined threshold, depending upon two parameters utilisation factor and strength of traffic load of the surrogates.

Keywords: content delivery network; CDN; utilisation factor; traffic load; population threshold; deployment cost; K-means clustering; replica server; flash crowd.

DOI: 10.1504/IJAIP.2023.132382

International Journal of Advanced Intelligence Paradigms, 2023 Vol.25 No.3/4, pp.219 - 233

Received: 11 May 2018
Accepted: 08 Jan 2019
Published online: 19 Jul 2023 *

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