Parallel processing using Raspberry Pi

• Main Author: Zhou, Lina
• Other Authors: Cai Wentong
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer systems organization::Performance of systems
• Online Access: http://hdl.handle.net/10356/59176
• Institution: Nanyang Technological University

A computer cluster refers to a set of loosely or tightly connected computers that solve computational problems together and in parallel. Raspberry Pi (RPi) Cluster is a set of Raspberry Pi (up to 64) connected to a switch or number of switches and server as a lightweight version of MPI for parallel processing. The purpose of this project is to determine the performance measurement by RPi cluster is able to replace the current cluster in NTU server that support one of the teaching materials in “CE/CZ 4011 Parallel Computing”. Firstly in this project, the student is required to conduct an experiment on RPi cluster (A master with 24 nodes). Based on the setup, perform a sample code to compute Pi (3.1415926) as well as several benchmarking such as NAS Parallel Benchmarks (NPB). In the project, the student would explain in details every individual step on how to setup the RPi cluster, make and run benchmarks and how to determine RPi’s speed up and efficiency by benchmark results.