Machine Learning Models for Anomaly Detection in Microservices
Vijay Ramamoorthi
Independent Researcher
Keywords: Microservice architectures, Anomaly detection, Machine learning, Time series models, AI-based monitoring, Distributed systems
Abstract
Microservice architectures have revolutionized the way distributed systems are designed, offering enhanced scalability, flexibility, and resilience. However, the complexity of managing numerous interconnected services poses significant challenges in ensuring system reliability and performance, particularly when detecting anomalies in real-time. Traditional monitoring tools often struggle with the high-dimensionality and dynamic nature of microservices. This paper presents a comprehensive evaluation of AI-driven techniques for anomaly detection in microservice environments, focusing on both supervised and unsupervised learning models, as well as time-series forecasting methods. Through an extensive analysis of Random Forest, Support Vector Machines (SVM), Autoencoders, Isolation Forest, ARIMA, LSTM, and Prophet models, we demonstrate their effectiveness in detecting performance anomalies across various system metrics such as CPU usage, memory consumption, network I/O, and latency. The results indicate that LSTM and Random Forest offer the highest precision and recall rates, while hybrid models combining multiple techniques present a promising avenue for improving detection accuracy. Our findings contribute to the growing body of research aimed at optimizing anomaly detection frameworks for microservice architectures, highlighting the importance of leveraging AI to address the evolving challenges of modern distributed systems.
Author Biography
Vijay Ramamoorthi, Independent Researcher
Vijay Ramamoorthi is a seasoned software architect with a background in artificial intelligence and machine learning. He has designed and implemented complex systems for Fortune 500 companies and has a passion for building scalable, reliable software solutions. His expertise spans cloud computing, microservices, and distributed systems. Vijay holds a Master's degree in Computer Science and a Bachelor's in Mathematics
