NOSQL DATABASES IN REAL-TIME SYSTEMS: PERFORMANCE, SCALABILITY, AND USE-CASE ANALYSIS

Abstract

The exponential growth of real-time data generated by digital ecosystems such as location-based transportation platforms, e-commerce, IoT sensor networks, real-time analytics, financial fraud detection, and intelligent manufacturing demands high-performance and horizontally scalable database solutions. Traditional relational database management systems (RDBMS), with rigid schemas, ACID-based consistency, and vertically scaled infrastructures, experience bottlenecks when faced with large and rapidly changing datasets. As a result, NoSQL (Not-Only-SQL) databases have emerged as a major alternative that provide flexible schemas, distributed processing frameworks, and optimized data models for high throughput and minimal latency computing environments. This paper provides a comprehensive evaluation of four leading NoSQL databases Redis, MongoDB, Cassandra, and Neo4j focusing on their architecture, performance, scalability, operational mechanisms, and suitability for real-time applications. Using a qualitative research methodology based on published peer-reviewed literature and industrial case deployment analysis, this study highlights that NoSQL systems outperform relational databases in realtime distributed environments but also face challenges regarding consistency management, security, and complexity of deployment. The study concludes with recommendations and future research directions aimed at improving the efficiency and reliability of NoSQL for real-time intelligent systems.