eprintid: 10613
rev_number: 2
eprint_status: archive
userid: 1
dir: disk0/00/01/06/13
datestamp: 2023-11-09 16:37:14
lastmod: 2023-11-09 16:37:14
status_changed: 2023-11-09 16:31:48
type: article
metadata_visibility: show
creators_name: Abdullahi, A.U.
creators_name: Ahmad, R.
creators_name: Nordin Zakaria, M.
title: Experimental performance analysis of B+-trees with big data indexing potentials
ispublished: pub
keywords: Advanced Analytics; Big data; Data Analytics; Forestry; Indexing (materials working); Indexing (of information), Analytical tool; Approach evaluation; B+-trees; Comparative experiments; Data platform; Experimental performance analysis; Indexing techniques; Relational Database, Trees (mathematics)
note: cited By 0
abstract: Indexing has contributed significantly to performance improvement of query processing. In doing that, many indexing techniques have been developed and deployed in the past few decades. Some of these techniques are static and based on data, while others are dynamic, adaptive and usually based on queries. With the emergence of big data and its analytical tools, the implementation of indexing is expected to extend to such tools for performance improvement. However, up until now very few of the successful indexing techniques in Relational databases have been explored on big data platforms. Hence, this paper shares the results of a comparative experiment conducted on B+-Tree variants in indexing big data. This is done with a view to expose the B+-trees indexing potentials in big data analytics. The selected trees are: Traditional B+-tree, Partitioned B+-tree, Simple prefix B+-tree and Hybrid Partitioned Simple Prefix B+-tree. In the experiment, factors for index approach evaluation are used to evaluate and compare the performance of the data structures. The experimental results showed that Partitioned B+-tree has in sum performed better in terms of execution time than the remaining B+-tree variants. This is beside the flexibility and adaptability of the Partitioned B+-tree. © Springer International Publishing AG 2018.
date: 2018
publisher: Springer Science and Business Media Deutschland GmbH
official_url: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090372707&doi=10.1007%2f978-3-319-59427-9_3&partnerID=40&md5=9434143cba3f0b8007c8939797924309
id_number: 10.1007/978-3-319-59427-9₃
full_text_status: none
publication: Lecture Notes on Data Engineering and Communications Technologies
volume: 5
pagerange: 20-29
refereed: TRUE
issn: 23674512
citation:   Abdullahi, A.U. and Ahmad, R. and Nordin Zakaria, M.  (2018) Experimental performance analysis of B+-trees with big data indexing potentials.  Lecture Notes on Data Engineering and Communications Technologies, 5.  pp. 20-29.  ISSN 23674512