Self-adapting real-time data ecosystems with autonomous multi-agent systems

Sai Kiran Reddy Malikireddy *

Independent Researcher, USA.
 
Research Article
World Journal of Advanced Research and Reviews, 2022, 13(03), 593-607
Article DOI: 10.30574/wjarr.2022.13.3.0227
 
Publication history: 
Received on 06 February 2022; revised on 20 March 2022; accepted on 23 March 2022
 
Abstract: 
As data ecosystems grow increasingly complex, traditional centralized control systems struggle to manage dynamic, large-scale workloads effectively. This paper introduces an autonomous multi-agent system (MAS) framework that uses reinforcement learning to enable self-adapting real-time data ecosystems. Each agent optimizes specific pipeline components, such as ingestion, transformation, and storage, while collaborating with other agents via a decentralized coordination protocol. Results from deployment in a smart city analytics platform demonstrate enhanced scalability and resilience, achieving a 40% improvement in system uptime and a 35% reduction in latency under dynamic workloads.
 
Keywords: 
Real-Time Data Processing; Autonomous Multi-Agent Systems; Self-Adaptive Systems; Data Ecosystem Optimization; Distributed Intelligence
 
Full text article in PDF: 
 

This paper has received CERTIFICATE OF BEST PAPER (Volume 13, Issue 3, Year 2022).

Download Certificate

 

Share this