AI Meets Load Balancing: A Reinforcement Learning Approach for Traffic Spike Resilience

Soubhik Roy; Khushi Ojha; Akash Dasandi

AI Meets Load Balancing: A Reinforcement Learning Approach for Traffic Spike Resilience

Author(s)	Soubhik Roy, Khushi Ojha, Akash Dasandi
Country	India
Abstract	Modern microservices architectures face increasing challenges in handling unpredictable traffic surges, often leading to performance degradation, latency issues, and inefficient resource utilization. Traditional load balancing techniques, such as Round Robin and Least Connections, rely on static rules that fail to adapt to real-time fluctuations, resulting in system bottlenecks and suboptimal performance. This research introduces AI-Driven Adaptive Load Balancing (AI-ALB), an advanced reinforcement learning-based system designed to dynamically manage traffic across cloud-native microservices environments. AI-ALB continuously monitors real-time traffic patterns, learns from historical data, and proactively adjusts its load balancing policies to optimize performance and prevent congestion. The system integrates Envoy Proxy for traffic management, TensorFlow/PyTorch for model training, and Kubernetes for orchestration, ensuring automated scaling without manual intervention. To assess AI-ALB’s real-world viability, we conducted extensive testing across AWS, Google Cloud, and Azure, demonstrating a 40% reduction in response latency and a 30% improvement in resource efficiency compared to traditional methods. Additionally, we optimized reinforcement learning training time using batch processing, transfer learning, and distributed computing, reducing learning overhead by 50%. Furthermore, we developed deployment strategies ensuring seamless integration with cloud-native services, including AWS Load Balancer, Google Cloud Load Balancing, and Kubernetes Ingress Controllers. These enhancements make AI-ALB a scalable, efficient, and resilient solution for managing dynamic cloud environments. By bridging AI with cloud infrastructure, this research paves the way for intelligent, self-learning load balancing in large-scale distributed systems.
Keywords	Adaptive Load Balancing, Reinforcement Learning, Microservices, Traffic Optimization, AI-driven Systems, Kubernetes, Envoy Proxy.
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 7, Issue 2, March-April 2025
Published On	2025-03-09

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 7 Isu 2 Cover Page Vol 7 Isu 1 Cover Page Vol 6 Isu 6

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

AI Meets Load Balancing: A Reinforcement Learning Approach for Traffic Spike Resilience

Share this