Investigation cloud data storage

Abstract

A cloud database is a database that typically runs on a cloud computing platform. Of the databases available on the cloud, traditional data model is SQL-based. The recent trend is to move on to NOSQL data model. Now, the question is which database approach is better to choose in this era of ‘Big Data’? SQL databases are difficult to scale, meaning they are not natively suited to a cloud environment, although cloud database services based on SQL are attempting to address this challenge. On the other hand, NOSQL databases are built to service heavy read/write loads and are able scale up and down easily, and therefore they are more natively suited to running on the cloud. Our aim for thesis is to investigate suitable data storage for cloud. Considering the ‘Big Data’ scenario of today’s world, we set forth to choose the NOSQL database model as the preferred solution for cloud computing. This paper aims to show two investigations on different branches of cloud data storage. The first analysis is based on the case study of performance benchmarking on 3 popular NOSQL databases - MongoDB, Cassandra, and HBase. The next part of investigation includes an experiment on the most popular ‘Big Data’ management framework – namely, Hadoop. Hadoop uses MapReduce for parallel computation, but writing MapReduce function is hard for programmers. So, our experiment is to configure HIVE data warehousing system on the top of Hadoop as a wrapper, so that end users gets benefit of using a SQL-like language, which is known as ‘HiveQL’ and provided by HIVE even if with the environment of complex MapReduce function.