Design and Development of Virtualized Hybrid Architecture for Cloud Computing

Abstract

Cloud Computing is a buzzword for IT infrastructure. There is a shift in business from desktops towards Cloud Computing. Cloud Computing is not a completely new concept; it has intricate connection to the established Grid Computing paradigm, and other relevant technologies such as utility computing, cluster computing and distributed systems in general. Cloud DBMS is a distributed database that delivers a query service across multiple distributed database nodes located in multiple data centres, including cloud data centers. The management of data centre is very costly and energy consuming. The cloud computing by providing the database- as- a-service helps the organizations in reduction of labour and infrastructure cost. In this thesis, the MapReduce and Parallel DBMS are compared focusing on Cloud DBMS properties, concluding that there is a need of a hybrid architecture which can combine the properties of these traditional approaches and fulfil the requirements of current Clouds. The Hybrid Architecture is implemented in virtualized environment using Hadoop Framework, Java for MapReduce programming and Mysql for query execution in Structured Query Language (SQL). The MapReduce is extended to provide homogeneity to work on relational databases.