Cloud System Enhancement using improved Workflow Task Ranking System

Abstract

There is not even iota of doubt that cloud computing has reached at its peak importance as compare to grid and parallel computing. In cloud systems, two types of algorithms can be imposed. These are heuristics and meta- heuristics. Heuristics algorithms like HEFT are applicable for the purpose of ranking of tasks but for better optimization, meta-heuristics algorithms like GA-PSO, hybrid optimization algorithms are more useful. Diverse systems use Scientific Workflows. Collections of multiple computational tasks are included in the workflow and scheduling of cloud computing science workflows is a demanding research area. A workflow-scheduling algorithm works in two stages. First, take input workflow tasks and rank the tasks based on parameters like makespan, cost and deadline constraints. During second stage, apply scheduling algorithm on ranked tasks and then schedule the tasks to cloud resources. In this paper, an improved workflow tasks ranking algorithm has been implemented. The findings after simulation show that our ranking algorithm has provided better results in terms of cost, time and deadline constraint over existing approaches.