Processing, Management Concepts, and Cloud Computing

This chapter deals with concepts behind the processing of big data such as parallel processing, distributed data processing, processing in batch mode, and processing in real time. There are basically two different types of data processing, namely, centralized and distributed data processing. Shared everything architecture is a type of system architecture sharing all the resources such as storage, memory, and processor. Shared‐nothing architecture is a type of distributed system architecture that has multiple systems interconnected to make the system scalable. The main purpose of virtualization in big data is to provide a single point of access to the data aggregated from multiple sources. Big data and cloud computing are the two fast evolving paradigms that are driving a revolution in various fields of computing. Cloud computing technology is broadly classified into three types based on its infrastructure: public cloud; private cloud; and hybrid cloud. Cloud storage adopts a distributed file system and a distributed database.