Building a computer cluster in SGU with apache hadoop mapreduce technology
The purpose of this research is to study the feasibility and effectiveness of using Apache Hadoop to a build computer cluster in SGU. The result of this research is expected to help SGU to determine whether the performance of a computer cluster could satisfy a high computational power requirement for the future research. In this research Apache Hadoop the software framework for building a computer cluster will be implemented on the existing resources in SGU. Once the computer cluster is built, the measurement of computing performance will be conducted using the official benchmark tools that included inside the Apache Hadoop package. The benchmarking process will be done on 2 difference machine. The first machine is the computer cluster itself while the second machine will be a computer with a high-end specification. The result of the benchmarking process will be used to compare the performance of a computer cluster against the performance of a single high-end computer. The study found that the performance of a computer cluster that was build using Apache Hadoop is greater than its competitor. Therefore the computer cluster is ready to be utilized for the research purposes.
No other version available