Title: ICRA: index based cache replacement algorithm for cloud storage

Authors: Yuwei Zhao; Tinghuai Ma; Yu Hao; Wenhai Shen; Yuan Tian; Abdullah Al-Dhelaan

Addresses: School of Computer Software, Nanjing University of Information Science Technology, Jiangsu, Nanjing 210-044, China ' CICAEET, Jiangsu Engineering Centre of Network Monitoring, Nanjing University of Information Science Technology, Jiangsu, Nanjing 210-044, China ' School of Computer Software, Nanjing University of Information Science Technology, Jiangsu, Nanjing 210-044, China ' National Meteorological Information Center, Beijing, 100-081, China ' Computer Science Department, College of Computer and Information Sciences, KingSaud University, Riyadh 11362, Saudi Arabia ' Computer Science Department, College of Computer and Information Sciences, KingSaud University, Riyadh 11362, Saudi Arabia

Abstract: Meteorological dataset is usually big and has a clear field. Due to its text form, each read and display will take up more system resources. In addition, the loading speed of the webpage also affects the user's online experience. According to these problems, we proposed a cache replacement algorithm combined with indexing algorithm called index cache replacement algorithm. The main innovation of this paper is that the index is created while reading the document and the page data at the first time and analyse, sort, and cache the indexing data at the same time. When users browse the same page, we can directly index file to query and return the corresponding data. Thus, it enhances the effectiveness of cached data and cache query and improves the query file and byte hit rate which finally improve the performance of the network.

Keywords: meteorological data; documents; pages; cache replacement algorithms; indexes; queries.

DOI: 10.1504/IJSNET.2019.097556

International Journal of Sensor Networks, 2019 Vol.29 No.1, pp.48 - 57

Received: 21 Jun 2018
Accepted: 12 Jul 2018

Published online: 28 Jan 2019 *

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