Scalability and Throughput Analysis: Benchmarking PostgreSQL, MongoDB, and CouchDB in Distributed Environments

Abstract

The rapid evolution of digital technologies has intensified the demands on modern web applications, particularly in managing high-volume, concurrent read and write operations. Traditional relational databases, which rely on vertical scaling, often face limitations in performance, scalability, and cost-efficiency when handling large-scale data. In contrast, NoSQL databases especially document-oriented systems—provide flexible, horizontally scalable solutions tailored for distributed environments. However, selecting the most efficient database system for optimal throughput and low latency remains a challenge due to the diverse architectures of available NoSQL options. This study presents a comparative performance analysis of one relational database (PostgreSQL) and two document-oriented NoSQL databases (MongoDB and CouchDB) under distributed system conditions. Using a custom benchmarking tool, we evaluate their read/write throughput, latency, and scalability to determine their suitability for high-demand applications. Our findings aim to guide developers in selecting the most appropriate database solution based on performance requirements.