Effective Cloudlet Scheduling Algorithm for Load Balancing in Cloud Computing Using Fuzzy Logic

Effective Cloudlet Scheduling Algorithm for Load Balancing in Cloud Computing Using Fuzzy Logic

Ali Wided, Numan Çelebi, Bouakkaz Fatima
Copyright: © 2023 |Pages: 18
DOI: 10.4018/979-8-3693-0593-5.ch010
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Clouds contain a huge number of virtualized resources that can be made available instantaneously. Cloud technology offers a wide array of services, encompassing platforms, hardware, and software, effectively providing almost anything as a service. A single host represents a physical computer component, while a datacenter comprises numerous hosts responsible for managing virtual machines throughout their life cycles. The efficient scheduling of virtual machine requests plays a crucial role in cloud, as it ensures that requested tasks are completed within the shortest time according to user-defined preferences. This chapter introduces an optimization model based on fuzzy logic for scheduling tasks in cloud computing. The proposed model was tested and evaluated using a fuzzy logic-based scheduling algorithm. The proposed algorithms were compared against scheduling algorithms without fuzzy logic. The experimental findings undeniably establish the superiority of the proposed algorithm.
Chapter Preview
Top

Background

In this section, we introduce several relevant works to our research. These studies address the challenges and issues associated with scheduling techniques and task scheduling in an academic context. They propose optimization solutions incorporating hybrid algorithms, with the latter approach utilizing fuzzy logic to attain optimal scheduling.

k-means clustering and Fuzzy logic are used in the work by Hamdani (Hamdani,2021) to present an improved Active VM load balancing solution that lowers datacenter transfer costs, total virtual machine expenses, datacenter processing times, and response times. The approach efficiently divides the load and improves CPU efficiency. The Cloud Analyst simulator is used to implement the algorithms. They are then compared against some load balancing strategies, such as Particle Swarm Optimisation (PSO), Round Robin (RRB), Active VM Load Balancing (ESP), Throttled algorithms (THR), Ant Colony Optimisation (ANT), Honey Bees load balancer (BEE), and Throttled algorithms (THR). The proposed algorithm clearly outperforms the competition in terms of overall performance, load balancing, and throughput capability. But, because bandwidth and other relevant aspects aren't considered, the study can't improve load balancing. An issue is that fuzzy sets and weights are computed so that when VMs operate equally well, the method converges to an equally distributed form.

Complete Chapter List

Search this Book:
Reset