Optimized multi‐objective Q‐learning with enhanced beetle swarm optimization based scientific workflows scheduling on cloud computing environment

Abstract

Summary Cloud technology has raised significant prominence providing a unique market economic approach for resolving large‐scale challenges in heterogeneous distributed systems. Through the use of the network, it delivers secure, quick, and profitable information storage with computational capability. Cloud applications are available on‐demand to meet a variety of user QoS standards. Due to the large number of users and tasks, it is important to achieve efficient planning of tasks submitted by users. One of the most important and difficult nondeterministic polynomial‐hard challenges in cloud technology is task scheduling. To overcome the problem, in this article, optimized multi‐objective Q‐learning with EBSO‐based task scheduling on the cloud is proposed. Q‐learning is one of the machine learning algorithms that can solve these types of problems. The proposed approach is divided into two processes: task prioritization and resource selection. In this article, initially, the tasks are prioritized by using the Q‐learning algorithm. After the prioritization, the tasks are assigned to resources by using the enhanced beetle swarm optimization (EBSO) algorithm. To achieve this, the multi‐objective function is designed based on makespan, cost, and resource utilization. The effectiveness of the suggested method is evaluated using a variety of criteria such as makespan, resource utilization, and cost.