EDLB-R: Enhancing Cloud Computing Performance Through Intelligent Load Balancing and Resource Allocation

Cloud Computing, Load Balancing, Task Scheduling, RAM-Aware Scheduling, Quality of Service

Meenakshi Saini1
Designation: Research Scholar Institution:
Galgotias University, Greater Noida
Department: Electrical, Electronics and
Communication Engineering
Email:meenakshisaini929@gmail.com
Address: Plot No. 2, Sector 17-A, Yamuna
Expressway, Greater Noida, Gautam Buddh
Nagar, Uttar Pradesh, India, PIN – 203201
ORCID-ID:https://orcid.org/0009-0006-5274-8784

Prabhakar Agarwal2 (Corresponding Author)
Designation: Associate Professor
Qualification: PhD Institution: Galgotias
University, Greater Noida Department:
Electrical, Electronics and Communication
Engineering
Email: prabhakarvirgo15@gmail.com
Address: Plot No. 2, Sector 17-A, Yamuna
Expressway, Greater Noida, Gautam Buddh
Nagar, Uttar Pradesh, India, PIN – 203201
ORCID-ID:https://orcid.org/0000-0003-4818-9351

Received:16-06-2026
Revised:20-07-2026
Accepted: 24-07-2026

European Journal of Prosthodontics and Restorative Dentistry (2026) 34(5s), 332–348

EDLB-R: Enhancing Cloud Computing Performance Through Intelligent Load Balancing and Resource Allocation

Cloud computing has become a widely adopted paradigm for providing scalable and on-demand computing resources. However, efficient task scheduling and load balancing remain major challenges due to dynamic workloads and heterogeneous virtual machine (VM) resources. Conventional load balancing algorithms primarily focus on processor utilization while often neglecting memory availability, leading to increased execution time, poor resource utilization, and uneven workload distribution. This paper proposes an Enhanced Dynamic Load Balancing with RAM Awareness (EDLB-R) algorithm that incorporates available RAM into the VM selection process to improve scheduling efficiency and resource allocation. The proposed algorithm was implemented using the CloudSim simulation toolkit and evaluated in the Eclipse IDE using 2 virtual machines with workloads ranging from 10 to 100 cloudlets. Performance was assessed using conventional metrics, including total makespan, average makespan, average execution time, resource utilization, and load balancing, along with additional Quality of Service (QoS) metrics comprising response time, throughput, and deadline success rate. Experimental results demonstrate that EDLB-R consistently outperforms the benchmark algorithm. At 100 cloudlets, the proposed EDLBR algorithm reduces total makespan by 29%, average makespan by 7%, and average execution time by 14% compared with the baseline EDLB algorithm. Furthermore, resource utilization improves by 2.01%, increasing from 65.57% to 66.89%, while load balancing is enhanced by 34%, increasing from 90.34% to 121.06%. Overall experimental results demonstrate that the proposed EDLB-R algorithm consistently outperforms the baseline across all evaluated workloads by reducing scheduling overhead, improving resource utilization, and achieving better workload distribution. These findings indicate that EDLB-R provides an efficient, scalable, and reliable scheduling solution for Infrastructure-as-a-Service (IaaS) cloud environments.

•••••••••••••••••••••••••••••••• ejprd.org- Published by Riset Publishing Services LLC.

EJPRD

Copyright © 2026 by Riset Publishing Services LLC