Big Data Platform as a Service for Anomaly Detection

Abstract

Multipurpose Computation is reached with the platform as a service with container orchestration (PaaSCO) which integrates containerized technologies in a package-based style. These platforms include Kubernetes, OpenShift, Distributed Cloud Operating System DC/OS, Cloud Foundry, and Docker Swarm. Using a PaaSCO facilitates and accelerates the construction of platforms for developing applications by unifying components and software frameworks to build applications for different use cases.Containers enable ubiquity, portability, and distributed computing capabilities, as well as the programming and development of different applications and services to engineering and sciences areas like the internet of things, software development, data analytics, microservices, and artificial intelligence. This means that the same solution can be deployed in different PaaSCO environments including public, private, and hybrid clouds to obtain the same service. The National Laboratory of Information Technologies of the Autonomous University of Ciudad Juárez (LaNTI) offers computing infrastructure services to researchers inside and outside the institution. Due to their lack of expertise in infrastructure and computing tools, most researchers have difficulty installing and configuring the software tools necessary for their research. We present a solution using the PaaSCO Distributed Cloud Operating System (DCOS) through the integration of a stack of components required by a Big Data architecture. By implementing the solution proposed in this work, we are contributing to the academic and research environment by accelerating the generation, and implementation of components required by the BigDataAnalytics system. Using containerized tools makes it easy for researchers to focus on the substantive part of their research. In addition to allowing data analytics tasks with a focus on anomaly detection, two experiments demonstrating resource elasticity and isolation are presented to demonstrate the platform additional benefits.