Performance Evaluation of JSON Libraries in Kotlin Server Environments: Jackson, Gson, and Kotlinx Serialization

Abstract

The widespread adoption of Kotlin for server-side development has emphasized the importance of efficient JSON processing. This study aimed to empirically evaluate and compare the performance of three JSON libraries—Jackson, Gson, and KotlinX Serialization—in Kotlin-based server environments. Using structured experiments, the research measured execution time, memory utilization, and CPU consumption during parsing, serialization, and deserialization tasks involving nested JSON data with incremental complexity. Statistical analyses confirmed significant performance differences among libraries. KotlinX Serialization consistently outperformed Jackson and Gson across execution speed, memory efficiency, and CPU utilization, primarily due to its compile-time code generation approach. Jackson showed superior memory management and efficient serialization for smaller datasets, while Gson exhibited lower overall performance. These results offer critical guidance for developers selecting appropriate JSON libraries, highlighting KotlinX Serialization's suitability for high-performance applications. The findings significantly enhance the current understanding of JSON processing efficiency and suggest avenues for further investigation into additional libraries and real-world application contexts.