Performance comparison between Hadoop and HAMR under laboratory environment

With the development of the Internet technology, data explosion is about to take place. To handle such enormous amount of data, including storing, organizing and analyzing, the capability of a single machine is far from sufficient. Therefore, it is meaningful to build a distributed computing platform for not only academic purpose, but also industrial usage. Hadoop is one of the most popular and developed solutions to Big Data. It provides reliable, scalable, fault-tolerance and efficient service for large scale data processing based on HDFS and MapReduce. HAMR is another new technology which is said that runs faster than Hadoop with less memory and CPU consumptions. This paper makes a performance comparison between Hadoop and HAMR based on running PageRank by measuring running time, maximum and average memory and CPU usage. The result can be helpful for constructing distributed computer platform.