Optimasi Performa API Aplikasi E-Commerce Menggunakan Redis Caching

Abstract

System performance is one of the important factors in the development of e-commerce applications that serve thousands of data requests every second. High intensity of user requests often leads to an overload of queries on the database, which has an impact on increased response times and decreased user experience. One of the effective solutions to overcome this problem is to implement a caching mechanism. This study aims to determine the effect of the use of Redis as a caching layer on improving the performance of Express.js-based APIs  in e-commerce applications. The method used was a quantitative experiment with three test scenarios, namely API without Redis, API with Redis in cold cache condition, and API with Redis in warm cache condition. The test was carried out using  the k6 load testing tool to simulate the user load of up to 100 virtual users, with the observed metrics including response time, throughput, and error rate. The test results showed that the implementation of Redis was able to reduce the average response time by up to 69% and increase throughput three times compared to the condition without Redis. The best performance is obtained in the warm cache condition, where Redis has stored data in memory and is able to provide a response with an average time of 32.68 ms. This study proves that Redis is an effective solution to optimize API performance in e-commerce systems with high and repetitive data access patterns.