Measuring performance quality scenarios in big data analytics applications

Abstract

Big data analytics (BDA) applications use advanced analysis algorithms to extract valuable insights from large, fast, and heterogeneous data sources. These complex BDA applications require software design, development, and deployment strategies to deal with volume, velocity, and variety (3vs) while sustaining expected performance levels. BDA software complexity frequently leads to delayed deployments, longer development cycles and challenging performance monitoring. This paper proposes a DevOps and Domain Specific Model (DSM) approach to design, deploy, and monitor performance Quality Scenarios (QS) in BDA applications. This approach uses high-level abstractions to describe deployment strategies and QS enabling performance monitoring. Our experimentation compares the effort of development, deployment and QS monitoring of BDA applications with two use cases of near mid-air collisions (NMAC) detection. The use cases include different performance QS, processing models, and deployment strategies. Our results show shorter (re)deployment cycles and the fulfillment of latency and deadline QS for micro-batch and batch processing.