A Fuzzy based Pathfinder Optimization Technique for Performance-Effective Task Scheduling in Cloud

Abstract

Cloud computing provides a shared pool of resources in a distributed environment. It supports the features of utility-based computing. Task scheduling is a largely studied research topic in cloud computing, which targets utilizing cloud resources for tasks by considering the objectives specified in QoS. Optimal task scheduling is an NP-hard problem, which is time-consuming to solve with precise methods and depends on many factors such as completion time, latency, cost, energy consumption, throughput, and load balance on the machines. Therefore, using meta-heuristic algorithms is a good selection. This paper uses the Pathfinder optimization Algorithm (PFA) for the task scheduling problem; but when the dimension of a problem is extremely increased, the performance of this algorithm decreases. In the last iterations, fluctuation rate (A) and vibration vector (e) converge to 0, and finding a new solution is impossible. We used fuzzy logic to overcome this shortcoming and named the new algorithm Fuzzy-PFA (FPFA). In this paper, makespan, energy consumption, throughput, tardiness, and degree of imbalance are considered as objective functions. Our goal is to minimize the makespan, energy consumption, tardiness, and degree of imbalance while maximizing throughput. Finally, different algorithms such as Firefly Algorithm (FA), Bat Algorithm (BA), Particle Swarm Optimization (PSO), and PFA are used for comparison. The experimental results indicate that the proposed scheduling algorithm can improve up to 34.2%, 16.2%, 15.9%, and 3.5% the objective function in comparison with FA, BA, PSO, and PFA, respectively.