Relix: A Comparative Latency Framework for REST and GRPC Using Redis Acceleration

Abstract

Microservice-based architectures require efficient communication mechanisms to ensure high performance and scalability. REST, which relies on HTTP/1.1 and JSON, is widely adopted due to its simplicity but suffers from larger payload sizes and higher latency. In contrast, gRPC utilizes HTTP/2 and Protocol Buffers, enabling faster communication through binary serialization and persistent connections. This research presents a comparative latency analysis of REST and gRPC microservices enhanced with Redis caching. Both systems were implemented using identical business logic and evaluated under two scenarios: without caching (cold requests) and with caching (warm requests). Key performance metrics such as latency and payload size were measured and visualized. The experimental results demonstrate that gRPC consistently outperforms REST in terms of latency, stability, and payload efficiency. Redis caching significantly improves performance in both approaches, with gRPC showing greater benefits due to its compact data representation. The study concludes that gRPC is more suitable for high-performance microservice communication, especially in latency-sensitive applications.