Study on SBC Cluster Computing Performance
| Autor: | CHANG,MIN-HAO, 張旻皓 |
|---|---|
| Rok vydání: | 2016 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 104 In recent years, computers, workstations, servers, embedded systems and mobile, scientific and technological progress very rapidly, and the rapid popularization of Internet, network speed upgrade very much, but with these advances produce large of data, so in the storage or use, this from the cloud computing is a good solution. Now, many data centers providing cloud computing and cloud services to solve the problem of big data in the world. we put forward setup clusters used Hadoop on the new Single-Board Computers Raspberry Pi, this Hadoop clusters provides computing and storage services, the burden of data centers construction is very small beacuse Raspberry Pi is low prices, purchase the server is much larger than SBC computer on resources, Hadoop running on each node using distributed computing MapReduce processing big data.we running pattern recognition on the SBC clusters and the single computer, they similar of price,Observed they performance,the SBC clusters better than single computer performance increase by about 20% and increase number of SBC can improve cluster processing speed in this research experimental data was apparently, and Hadoop used by Hadoop Distributed File System(HDFS) in the data storage, this security better than single computer. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|