Performances Evaluation of a Novel Hadoop and Spark Based System of Image Retrieval for Huge Collections

A novel system of image retrieval, based on Hadoop and Spark, is presented. Managing and extracting information from Big Data is a challenging and fundamental task. For these reasons, the system is scalable and it is designed to be able to manage small collections of images as well as huge collectio...

Full description

Saved in:
Bibliographic Details
Main Authors: Luca Costantini, Raffaele Nicolussi
Format: Article
Language:English
Published: Wiley 2015-01-01
Series:Advances in Multimedia
Online Access:http://dx.doi.org/10.1155/2015/629783
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849684522226941952
author Luca Costantini
Raffaele Nicolussi
author_facet Luca Costantini
Raffaele Nicolussi
author_sort Luca Costantini
collection DOAJ
description A novel system of image retrieval, based on Hadoop and Spark, is presented. Managing and extracting information from Big Data is a challenging and fundamental task. For these reasons, the system is scalable and it is designed to be able to manage small collections of images as well as huge collections of images. Hadoop and Spark are based on the MapReduce framework, but they have different characteristics. The proposed system is designed to take advantage of these two technologies. The performances of the proposed system are evaluated and analysed in terms of computational cost in order to understand in which context it could be successfully used. The experimental results show that the proposed system is efficient for both small and huge collections.
format Article
id doaj-art-c11ef158e09d46fab35fa9ae0609f48b
institution DOAJ
issn 1687-5680
1687-5699
language English
publishDate 2015-01-01
publisher Wiley
record_format Article
series Advances in Multimedia
spelling doaj-art-c11ef158e09d46fab35fa9ae0609f48b2025-08-20T03:23:26ZengWileyAdvances in Multimedia1687-56801687-56992015-01-01201510.1155/2015/629783629783Performances Evaluation of a Novel Hadoop and Spark Based System of Image Retrieval for Huge CollectionsLuca Costantini0Raffaele Nicolussi1Fondazione Ugo Bordoni, Viale del Policlinico, 147 Roma, ItalyFondazione Ugo Bordoni, Viale del Policlinico, 147 Roma, ItalyA novel system of image retrieval, based on Hadoop and Spark, is presented. Managing and extracting information from Big Data is a challenging and fundamental task. For these reasons, the system is scalable and it is designed to be able to manage small collections of images as well as huge collections of images. Hadoop and Spark are based on the MapReduce framework, but they have different characteristics. The proposed system is designed to take advantage of these two technologies. The performances of the proposed system are evaluated and analysed in terms of computational cost in order to understand in which context it could be successfully used. The experimental results show that the proposed system is efficient for both small and huge collections.http://dx.doi.org/10.1155/2015/629783
spellingShingle Luca Costantini
Raffaele Nicolussi
Performances Evaluation of a Novel Hadoop and Spark Based System of Image Retrieval for Huge Collections
Advances in Multimedia
title Performances Evaluation of a Novel Hadoop and Spark Based System of Image Retrieval for Huge Collections
title_full Performances Evaluation of a Novel Hadoop and Spark Based System of Image Retrieval for Huge Collections
title_fullStr Performances Evaluation of a Novel Hadoop and Spark Based System of Image Retrieval for Huge Collections
title_full_unstemmed Performances Evaluation of a Novel Hadoop and Spark Based System of Image Retrieval for Huge Collections
title_short Performances Evaluation of a Novel Hadoop and Spark Based System of Image Retrieval for Huge Collections
title_sort performances evaluation of a novel hadoop and spark based system of image retrieval for huge collections
url http://dx.doi.org/10.1155/2015/629783
work_keys_str_mv AT lucacostantini performancesevaluationofanovelhadoopandsparkbasedsystemofimageretrievalforhugecollections
AT raffaelenicolussi performancesevaluationofanovelhadoopandsparkbasedsystemofimageretrievalforhugecollections