Migrating Monolithic to Microservices: Comparative Performance Testing Evaluation Between Architectures

Abstract

Purpose: Various organizations and industries have migrated their systems from previously adopting monolithic architecture to microservices architecture. One of the advantages of microservices architecture that is desired to be achieved from the migration process is the performance side. Therefore, this research aims to conducted performance testing on the system that was migrated from monolithic to microservices in the previous study.

Methods: This research was conducted in several stages, such as designing and implementing software and applications, creating performance testing scenarios, executing scenario testing with load testing, stability testing (soak testing), and stress testing such as load testing, soak testing, and stress testing, and finally analyzing and reporting testing results using performance indicator in terms of response time, throughput, and error rate.

Result: The test results showed a significant increase in performance before and after the migration of the monolithic system to microservices. Application response time became faster, more requests could be handled, and the failure rate experienced by the system was smaller. This shows that system performance is better with the implementation of microservices architecture.

Novelty: This research presents a novelty in the form of a comparative evaluation of real deployment-based system performance between Laravel monolithic architecture and Golang gRPC-based microservices on the same application, with a seven-stage performance testing approach and the use of in-depth quantitative metrics using Apache JMeter.