Trustworthy Hybrid Cloud Architecture for Big Data Processing

Abstract

Big data refers to huge volume of data generated frequently from several domains such as internet, health care systems, weather monitoring systems, enterprises, sensors, and from private and public sectors. Big data are a priceless source of information and is structured, unstructured, or in semistructured form. Due to volume, variety, velocity, and veracity of big data, it is diand#64259;cult to process it using traditional systems. There is a need for data intensive computing platform to eand#64259;ciently store and process data. Big data analytics refers to the process in which valuable hidden data patterns are uncovered and presented as reports that helps in decision making and prediction. Big data analytics has received sizeable attention since it oand#64256;ers a great opportunity to uncover potentials from heavy amounts of data.It necessitates the use of mining algorithms and strong supporting frameworks. newlineCloud computing is considered as on-demand information technology that provides software, platform, and software as services. Big data necessitates the usage of cloud computing because of its flexibility, scalability, and data sharing capabilities. Cloud deployment models can be public, private, and hybrid cloud types. Public cloud can be used on-demand and charges are applied based on usage model. Private cloud is meant for an organization which provides more security to data. Hybrid cloud is the combination of private and public cloud and helps in data distribution and parallel processing. Hadoop MapReduce framework helps in execution of processing mechanisms by means of map and reduce operations. Apache Spark provides better performance than Hadoop because of its in-memory processing capability and uses cluster based execution of jobs that produce results quickly. Cloud computing is preferred for big data analytics as it provide not only unlimited storage but also supports eand#64259;cient data processing capability. An important basis of big data analysis and management is the availability of high-quality, precise, and trustwor