Diseño e implementación de microservicios para generar un modelo de análisis de transacciones masivas de tarjetas de débito para una institución financiera

Abstract

This work presents a proposal for the generation of microservices to capture the information of massive transactions of debit cards of a financial institution from a relational database, generate the respective process of extraction, transformation and load (ETL) and take it towards a database of non-relational data that allows through business intelligence mechanisms to convert stored data into timely and valuable information for decision-making. For the methodology, three phases were used: the first uses a mechanism within the relational database to detect the changes produced (CDC); the second, the design and implementation of a consumer in charge of listening to the messages and generating the respective process of transformation and loading to the non-relational database ensuring fault tolerance, reliability and scalability. Finally, through defined indicators, a control panel is designed that offers managers the necessary guidelines to generate decisions, measure the achievement of objectives and goals, thus offering, thanks to this visual tool, converting data into knowledge that generates competitive advantage in management and operational processes of the institution. The integration shows as a result the design of an architecture that efficiently manages thousands of records in real time, allowing at the same time greater speed, performance and optimization in searches of large volumes of data.