The advent of microservice architecture has brought unprecedented success in the banking sector. In this present era, it is imperative for banking sector to possess adequate preparedness to effectively manage fluctuations in transaction volumes during peak periods due to scalability changes. This paper presents a methodological approach to solving the problem of scalability in banking operations using Cameroon context. Microservice architectures, with their modularity and scalability, have proven to be a relevant answer to these challenges. This research explores common problems encountered in microservices, proposes a systematic methodology to address them, and illustrates these concepts through the development of a banking platform integrating modern technologies such as Docker, Docker Compose, and Kubernetes. Furthermore, load balancing techniques were examined, which were essential to optimize the performance of the banking application, and their impact on the efficiency of microservices. Crossover and mutation operators of the Adaptive Genetic Algorithm(AGA) were adopted to avoid premature convergence and to minimize plateau, which enhanced the diversity of population evolution and effectively reduced data transmission time between banking services. The development environment was properly set up to support the research goals. This includes ensuring the availability of essential tools such as Visual Studio Code, Eclipse, Java Development Kit(JDK), Apache Maven, Docker Desktop, Rester, Hey, RabbitMQ, Spring Boot, Actuator, Spring Cloud Gateway, Spring Cloud Config, Eureka, H2, MariaDB. The Hey tool was used for request tracing among microservices. To deploy this solution, modern technologies such as Docker, Docker Compose, and Kubernetes were employed which allowed efficient container management and service orchestration. Finally, performance and scalability tests were performed using the Hey tool, in order to evaluate the efficiency of our architecture and interpret the results for possible improvements. For performance testing analysis, four metrics were used that indicated satisfactory performance. The total response time averaged 0.6156 seconds, with the fastest response at 0.0056 seconds and the slowest at 0.2388 seconds. The average response time was 0.0545 seconds, achieving a throughput of 1624.5330 requests per second. A histogram analysis indicated that the majority of requests (406) fell within a response time of 0.052 to 0.076 seconds, confirming the system's overall efficiency. Latency distribution analysis showed that 50% of requests had latency under 0.0499 seconds, while 90% were below 0.0999 seconds, and 99% 0.1999 in seconds. No significant bottlenecks were identified in the various process steps, including Domain Name System (DNS) resolution and response handling.

Scalability Optimization Orchestration Microservice architecture Event-Driven-Controller Adaptive Genetic Algorithm