Big data process advancement

Abstract

Information in this era is thriving to be maintained on a verity of sources. Data is available in different patterns and forms. Combining and processing all different types of datasets in a heterogeneity database is near to impossible, specifically, if the information is moving and changing on many different sources on a continuous basis. Information is represented in different modules and nowadays processing data from various sources can lead to critical risk assessment results. Big Data is a concept introduced to cover the use of different techniques serving the desired goals by processing the given information. Processing huge amount of data is a big challenge for a single machine to perform, in this paper we will discuss this idea and demonstrate a module of clustered machines to work as a single entity towards achieving the desired tasks while working on parallel cohesively. The idea of a solution to combine different machines of different specification processing and power in a single cluster and then distributing input data of various data fairly to most powerful processing and well-designed data type machine in the cluster. Distribution of input data and storing mechanism will depend on machine specification, data processing, the power of a machine, balance loading and data type. We present our suggestion solving method by using Event-B based approach, the Key features of Event-B are the use of set theory as a modelling notation and we propose using the Rodin modelling tool for Event-B that integrates modelling and proving. © Springer International Publishing AG 2017.