FADTESE: A framework for automated deployment and effectiveness evaluation for big data tools

Abstract

Manual deployment of big data tools such as Hadoop, Sqoop, and Python is often slow, complex, and error prone because of extensive configuration steps, dependency conflicts, and inconsistent command-line execution. These challenges lead to unreliable installations and variations across systems. This study introduces framework for automated deployment and time, error, satisfaction evaluation (FADTESE), a unified framework that automates the installation of big data tools and evaluates its performance. The framework consists of two integrated components. The first is the automated deployment model, which validates environment readiness using the automation deployment readiness index (ADRI) and achieved a readiness value of 1.0 in this study. The second is the time, error, and satisfaction evaluation model, which quantifies improvements gained from automation and produced a score of 0.5941 through bootstrap resampling with ten thousand samples, indicating moderate effectiveness. The FADTESE script was technically validated across multiple Linux environments, including Ubuntu, Linux Mint, and AWS Ubuntu server systems. The performance evaluation involving eighty IT practitioners was conducted on Ubuntu systems to ensure consistent testing conditions and confirmed substantial gains in installation time, error reduction, and user satisfaction. Combining readiness and effectiveness yields a composite score of 0.5941 or 59.41%. FADTESE provides a reproducible and data driven method that standardizes big data deployment and improves reliability across local and cloud-based Linux environments.