A Survey on Multi-Objective Tasks and Workflow Scheduling Algorithms in Cloud Computing

A Survey on Multi-Objective Tasks and Workflow Scheduling Algorithms in Cloud Computing

Rajeshwari Sissodia, Man Mohan Singh Rauthan, Kanchan Naithani
Copyright: © 2022 |Pages: 16
DOI: 10.4018/IJCAC.297100
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

The challenge in cloud services is scheduling and allocating resources due to the exponential growth in demand and diversity of cloud resources. Scheduling is to allocate tasks across cloud resources so that scheduling algorithms reduce power consumption and offer cloud providers maximum return by reducing execution time. Various QoS parameters (such as makespan, load balancing, costs, etc.) are considered for efficient scheduling to reduce workload and enhance performance. Through this framework, multi-objective scheduling is a decision-making problem with multiple attributes considering the trade-off between the conflicting and competing parameters mentioned in the SLA between users and providers. This paper summarizes various multi-objective scheduling algorithms that consider contradictory and competing parameters or constraints to be optimized simultaneously. These algorithms are finally tabulated, presenting their advantages and disadvantages with cloud simulation tools and other QoS related parameters.
Article Preview
Top

Introduction

Cloud computing is the core of decentralized networks, grid computing and customer-centered architecture application development on the Internet. The Internet can be used as an accessible resource to allow users to provide and consume services like storage, platform, memory, software, hardware and infrastructure without physical assets. A cloud is a collection of virtual servers connected to a set of computers that are dynamically viewed as a data center. Cloud computing is an evolving and desirable choice for companies and organizations to address their day-to-day growing needs. The service provides dynamic, scalable services. Virtualization defines computing platforms, storage facilities and computations which are given to users rapidly in payment for what they utilize. Cloud Storage is a service offered by a provider that stores and retrieves data, space, computation, etc. Cloud storage provides three main services. SaaS, PaaS and IaaS. The SaaS is a software subscription that is offered to the client. The client needs a browser to access those applications stored in the Cloud. Google Apps are available to both organizations and individuals. The PaaS includes an infrastructure where the client does not control resources but deploys the application on the cloud platform. The IaaS offers networks, bandwidth, memory and other storage resources for applications. Users can also monitor operating systems, storage and applications deployed. Amazon Web Services provides IaaS for pay peruse. Customers can access various resources by using the new services, making the resources available. Customers can pay for services and thus the term "pay-per-usage" refers to a new service paradigm for the Cloud. Although cloud services are used, optimal scheduling is important because users rely directly on it to use resources. To solve the scheduling problem, this paper provides a study of multi-objective scheduling algorithms and their advantages and disadvantages. We hope our work is a starting point for new research in cloud computing and helps create better scheduling algorithms. The information below includes the list of abbreviations and description that can be found in table 1.

Complete Article List

Search this Journal:
Reset
Volume 14: 1 Issue (2024)
Volume 13: 1 Issue (2023)
Volume 12: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 4 Issues (2016)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing