A heuristic task scheduling algorithm in cloud computing environment: an overall cost minimization approach

With the advancement of the cloud computing environment, the users’ expectations to gain better services significantly increased. One of the most prominent parts of cloud systems is the task scheduling concept in which its improvement can increase the users’ satisfaction as a consequence. Most of th...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Cluster computing Ročník 28; číslo 2; s. 137
Hlavní autori: Boroumand, Ali, Hosseini Shirvani, Mirsaeid, Motameni, Homayun
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: New York Springer US 01.04.2025
Springer Nature B.V
Predmet:
Algorithms
Cloud computing
Computer Communication Networks
Computer Science
Customer relationship management
Customer services
Failure
Heuristic
Heuristic task scheduling
Literature reviews
Multiple objective analysis
Operating Systems
Optimization
Optimization models
Processor Architectures
Quality of service
Reliability
Scheduling
Search algorithms
Software services
User satisfaction
Workflow
Task scheduling
Cloud computing
Multi objective heuristic algorithm
Reliability
ISSN:1386-7857, 1573-7543
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:With the advancement of the cloud computing environment, the users’ expectations to gain better services significantly increased. One of the most prominent parts of cloud systems is the task scheduling concept in which its improvement can increase the users’ satisfaction as a consequence. Most of the published literature in this domain is extended to either a single-objective or bi-objective perspective. This paper presents a heuristic task scheduling algorithm for the optimization of makespan -cost-reliability (TSO-MCR) objectives. In addition, the users’ constraints are considered in the proposed optimization model. To this end, the task ranking approach, ignoring the unreliable processors, using Pareto dominance, and crowding distance approaches are utilized so the trade-off amongst potentially conflicting objectives is gained. To verify the effectiveness of the proposed TSO-MCR, its performance is compared with Multi-Objective Heterogeneous Earliest Finish Time (MOHEFT), Cost and Makespan Scheduling of Workflows in the Cloud (CMSWC), Hybrid Discrete Cuckoo Search Algorithm (HDCSA), and Multi-Objective Best Fit Decreasing (MOBFD) approaches. Since the comparative algorithms are bi-objectives, the multi-objective version of each is customized commensurate with the stated problem to prepare the same conditions. The simulation results prove that the proposed TSO-MCR significantly outperforms other state-of-the-art. It has 4.23, 8.93, 2.08, and 4.24% improvement against all counterparts in all 12 scenarios respectively in terms of makespan , total monetary cost, reliability, and the final score function incorporating all weighted objectives. It is worth mentioning that the comparison has been done on the datasets including both scientific workflow and random applications with different communication-to-computation ratio (CCR) values.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1386-7857
1573-7543
DOI:10.1007/s10586-024-04843-3