Using Redis for Caching Optimization in High-Traffic Web Applications

The relevance of this study lies in the growing demand for high-load web applications, especially in e-commerce, social media, and streaming platforms, where performance, stability, and scalability are crucial. As user loads increase, traditional relational database encounter performance bottlenecks, highlighting the need for efficient caching solutions. Redis, a high-performance in-memory key-value store, is frequently used in such scenarios; however, the impact of different caching strategies on its performance remains understudied. The purpose of this article is to comprehensively evaluate the effectiveness of Redis as a caching tool for optimizing web application performance. The experimental design involved testing a web application backed by PostgreSQL under four conditions: no cache, Redis with cache-aside, Redis with write-through, and Redis Cluster. User loads of 1,000, 5,000, and 10,000 were simulated using Locust, and performance metrics were collected through Prometheus. Statistical analysis was performed using a t-test, and results were visualized with graphs and tables. The results show that Redis significantly decreases average response time (e.g., from 1146 ms to 323 ms in cache-aside mode), increases throughput (up to 226 requests/sec), and reduces the load on the main database. Cache-aside proved most effective for read-intensive workloads, while Redis Cluster offered better stability under high concurrency. The findings confirm Redis as a valuable component for high-load applications. Future research should explore Redis in distributed database settings, compare it to emerging tools like KeyDB, and examine its energy efficiency in cloud environments.