Adaptive Cloud Resource Management using IBI Logic and Learning-based Optimization

Scheduling large, diverse, and dynamically arriving tasks effectively while preserving performance, resource usage, and Service Level Agreement (SLA) compliance is becoming more and more difficult in cloud computing environments. Drawing inspiration from the human mind’s adaptive and learning-driven characteristics, this paper presents the ILA-ML Scheduler, a novel cloud task scheduling framework that combines predictive machine learning models with the Incomprehensible but Intelligible-in-time (IbI) Logics Algorithm (ILA). The suggested scheduler makes use of Long Short-Term Memory (LSTM) networks to forecast workload dynamics and XGBoost to predict task execution time. The ILA optimization process uses these predictions as a guide to make proactive, flexible, and human-inspired task allocation choices. When compared to traditional scheduling algorithms like Min-Min, Round Robin, and Genetic Algorithm-based schedulers, extensive tests using a workload taken from the Google Cluster Workload Trace show that the ILA-ML Scheduler dramatically lowers SLA violation rates, improves virtual machine utilization, and reduces makespan. The outcomes demonstrate how well IbI logic-driven optimization and predictive machine learning work together to create next-generation cloud scheduling systems.