Designing parallel computing using raspberry pi clusters for IoT servers on apache Hadoop

The community has widely used the IoT tool because it makes it easy for users in various sectors. The implementation of IoT does not only require connectivity and devices but also requires a server. Servers that are accessed by many devices will have more and more computing processes on the server....

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2020-04, Vol.1517 (1), p.12070
Main Authors: Nugroho, S, Widiyanto, A
Format: Article
Language:English
Subjects:
Clusters
Physics
Servers
Software
Technology utilization
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The community has widely used the IoT tool because it makes it easy for users in various sectors. The implementation of IoT does not only require connectivity and devices but also requires a server. Servers that are accessed by many devices will have more and more computing processes on the server. The computing process on the server requires a high-performance but small-sized server device that is easy to connect with other devices. Raspberry Pi, known as an IoT server, is a computer with a small size, cheap, and very energy efficient. The parallel machine was built based on Raspberry Pi using parallel computing technology to have high performance. Nowadays, one of the parallel computing tools is Apache Hadoop. This research using clusters method with four pieces of Raspberry Pi with parallel computing technology using Apache Hadoop tools. Testing is done using Siege tools. The number of requests that used to test web server performance in this research was 10, 50, 100, and 150. The most optimal results are on requests by 50 with an increase of about 260% from a single node to cluster in the measurement of transactions, transaction rates, and throughput. Testing with 150 requests can only increase by 50% to 60%.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/1517/1/012070