An Efficient Meta-Heuristic Task Scheduling and Resource Constraints Based Machine Learning Framework on Different Scientific Workloads

Resumen

Multi-level task scheduling and resource optimization plays an essential role in the real-time
cloud computing environment. In the traditional multi-task meta-heuristic models are
independent of task prediction for different cloud servers and its resources. In this work, a
hybrid multi-task meta-heuristic based machine learning framework is proposed using different
workloads and server types. Experimental results show that the proposed multi-task metaheuristic
based machine learning approach has better efficiency than the conventional task
scheduling and resource constraints with different cloud server types.