Discrete Symbiotic Organisms Search Based Data-Intensive Scientific Workflow Optimization in Cloud Computing Environment

Abstract

Introduction

The ability to access and use computer resources has been totally transformed by cloud computing, which provides networking, processing, and storage capabilities as needed. As cloud-based applications and services have become more popular, so too have the quantity and complexity of tasks that require efficient scheduling. Scheduling tasks and locating data in cloud systems have become crucial concerns for scientific processes that significantly depend on data placement and tasks computing. With efficient scheduling, the best use of resources is ensured while overall execution time and communication delays are decreased. As an NP-complete problem, workflow scheduling is difficult to tackle optimally with conventional techniques and is computationally expensive.  Researchers are progressively tackling these problems by employing intelligent optimization methods based on natural occurrences.

Objectives

The goal of this work is to improve workflow scheduling performance in cloud systems by creating an optimization technique inspired by biology. The main goal is to reduce two important performance indicators: workflow completion time and data transfer time. The suggested method aims to enhance overall resource utilization, lower execution cost to minimize expenses, and speed up scientific computations in the cloud by effectively assigning jobs to virtual machines.

Methods

In order to manage the discrete character of cloud workflow scheduling, this study suggests using a Discrete Symbiotic Organism Search (DSOS) algorithm, which is a version of the symbiotic organism search technique. Iteratively evolving task assignments to generate near-optimal scheduling solutions. The approach is put into practice and tested in the popular simulation toolkit CloudSim, which is used to model and assess cloud computing infrastructures. To confirm the DSOS algorithm's efficacy and efficiency in resolving the scheduling issue in dynamic cloud settings, it is contrasted with a number of well-known scheduling techniques.

Conclusions

Workflow scheduling in cloud computing is a challenging problem that the suggested Discrete Symbiotic Organism Search algorithm successfully resolves. It improves scientific computation performance and helps create more responsive cloud services by cutting down on execution time, execution cost and placing data optimally.