DEVELOPMENT OF A BATCH PROCESSING APPLICATION WITH ZERO-DOWNTIME DEPLOYMENT

In today’s era of rapidly advancing software and information technologies, many operational processes involve repetitive tasks, requiring employees to routinely spend time completing them. Such situations significantly reduce work efficiency and lead to wasted time. This study aims to increase operational efficiency and enable employees to focus on more strategic and creative tasks. To this end, a batch processing system has been developed that is easy to use, automates routine operational and software processes, and provides uninterrupted service. This system is built on an architecture of admin and agent modules to coordinate and manage workloads. Angular and TypeScript have been used for front-end, while Python, FastAPI, SQLAlchemy, and PostgreSQL technologies have been used for back-end. Development and packaging tools such as Jenkins and Docker have been utilized. Sonarqube integration has been preferred for static code analysis. A hybrid transition architecture combining "Graceful Shutdown" and "Missed Job Recovery" approaches has been designed for uninterrupted service delivery and data security. With the developed system, processes have been made more effective and efficient through the automation of planned workflows. In this way, employees have been able to focus their time and energy on high value-added tasks, thus significantly increasing workforce productivity.