Year: 2025
Type: article
Abstract: The increasing use of Kubernetes has brought substantial operational complexity because manual management of its numerous dynamic components (pods, nodes, networks) is slow, error-prone, and unsustainable at scale. This research investigates how AIOps (Artificial Intelligence for IT Operations) principles can move past native automation to establish fully autonomous Kubernetes management. The proposed framework uses machine learning to detect anomalies, identify causes, and predict scaling needs before executing automatic remediation steps. Our methodology demonstrates that AIOps can enhance system reliability and reduce operational Toil while optimizing resource efficiency through closed-loop observation-action cycles, leading to self-healing Kubernetes ecosystems that require minimal human intervention.
Author: Kishan Raj Bellala
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4414557200