Optimized Task Scheduling and VM Allocation in Cloud Computing Using PPMMcNE and RSMBO Algorithms

This paper presents an optimized approach for task scheduling and virtual machine (VM) allocation in cloud computing environments, leveraging two novel algorithms. The proposed Phasmatodea Population Modified McNaughton Evolution (PPMMcNE) algorithm enhances the Phasmatodea Population Evolution (PPE...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Informatica (Ljubljana) Ročník 49; číslo 26; s. 181 - 199
Hlavní autori: Sanjay, Nisha, Sreedharan, Sasikumaran
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Ljubljana Slovenian Society Informatika / Slovensko drustvo Informatika 03.07.2025
Predmet:
Business metrics
Cloud computing
Completion time
Energy conservation
Energy consumption
Energy efficiency
Evolutionary algorithms
Genetic algorithms
Integrated approach
Optimization
Optimization techniques
Performance evaluation
Resource utilization
Response time
Scheduling
Task scheduling
Virtual environments
Workloads
ISSN:0350-5596, 1854-3871
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:This paper presents an optimized approach for task scheduling and virtual machine (VM) allocation in cloud computing environments, leveraging two novel algorithms. The proposed Phasmatodea Population Modified McNaughton Evolution (PPMMcNE) algorithm enhances the Phasmatodea Population Evolution (PPE) method by integrating Modified McNaughton's rule to generate a high quality initial task schedule and minimize delays. Complementarily, the Rat Swarm Modified Brucker Optimization (RSMBO) algorithm is introduced to refine VM allocation by reducing migration overhead and lowering energy consumption. The methods aim to optimize key cloud performance parameters- including turnaround time, waiting time, completion time, response time, makespan, cost, load balancing, and energy efficiency-thereby enhancing overall resource utilization and fairness. Comprehensive computational experiments were performed in Matlab using the publicly accessible G°CJ dataset, which comprises one month of resource utilization data, recording 123 million incidents across 1250 computers. The proposed method achieves a throughput of 0.942, exhibits a minimal task scheduling delay of 58.22 milliseconds, and maintains a queue waiting time of 43.66 milliseconds-all while reducing energy consumption to an average of 120 joules per task. Furthermore, energy consumption was quantitatively evaluated, with RSMBO consistently demonstrating significant reductions in energy usage compared to traditional baselines. These results validate that the integrated approach of PPMcNE and RSMBO offers superior scalability and efficiency, making it highly suitable for dynamic and large-scale cloud environments.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0350-5596
1854-3871
DOI:10.31449/inf.v49i26.7970