Genetic-Based Multi-objective Task Scheduling Algorithm in Cloud Computing Environment

Emara, Farouk A.

doi:10.22266/ijies2021.1031.50

Cited by 7 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Farouk A. Emara et al [33] introduced the G-MOTSA (Multi-Objective Task Scheduling Algorithm based on GA) to address resource management challenges in cloud. This approach was based on a modified GA, aims to optimize task scheduling by selecting the most suitable VMs for performing tasks and deploying them on appropriate servers.…”

Section: Literature Surveymentioning

confidence: 99%

A Modified Genetic-Based solution for Power-Aware Placement of Virtual Machines

Panwar,

Rauthan,

Barthwal

2024

Preprint

View full text Add to dashboard Cite

Cloud computing has developed as a ubiquitous technology for delivering services like storage, computing, etc. via the Internet. With the rising demand by customers for cloud computation and associated services, cloud service providers are developing various approaches that enhance the performance, reliability, and availability of cloud systems. Cloud computing uses virtualization to optimise resource usage and minimise power utilisation in data centers (DC). Efficient virtual machine (VM) placement strategies are crucial, especially when using advanced genetic techniques. This research paper introduces the use of a genetic meta-heuristic approach, named PowerGA, to optimise the integration of virtual machines (VM) onto the least number of physical machines (PMs) in cloud DCs. PowerGA optimises VM deployment in cloud DCs to minimise energy utilisation and Service Level Agreement (SLA) breaches, considering factors such as VM migration, host shutdown, overload count, and active physical machines. Extensive simulations using real workload data showed significant improvements over traditional strategies like PABFD, with PowerGA achieving a 25% reduction in energy consumption (EC), 43% fewer VM migrations, a 58% improvement in SLA violations, and a 72% reduction in host shutdowns over ten days of data from PlanetLab. These results highlight PowerGA's effectiveness in energy management and SLA enhancement, demonstrating the benefits of a meta-heuristic genetic algorithm in optimising VM placement for cloud computing efficiency.

show abstract

Section: Literature Surveymentioning

confidence: 99%

A Modified Genetic-Based solution for Power-Aware Placement of Virtual Machines

Panwar,

Rauthan,

Barthwal

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…In [86], another task scheduling algorithm, which consider resource management, employed a modified GA. This approach produced better results than Energy-aware Taskbased Virtual Machine Consolidation (ETVMC), Traveling Salesman Approach for Cloudlet Scheduling (TSACS), and ACO in terms of load balancing, makespan, resource utilization, throughput, and scheduling length.…”

Section: ) Genetic Algorithm (Ga) Based Techniquesmentioning

confidence: 99%

A Review on Task Scheduling Techniques in Cloud and Fog Computing: Taxonomy, Tools, Open Issues, Challenges, and Future Directions

Khan,

Aziz,

Osman

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Efficient task scheduling on the cloud is critical for optimal utilization of resources in data centers. It became even more challenging with the emergence of 5G and IoT applications that generate massive number of tasks with stringent latency requirements. This gives birth to fog/edge computing -a complementary layer to cloud. Tasks latency in fog computing can be reduced as processing in the network is done closer to the end devices and users, but every task cannot be scheduled in the fog due to limited resources availability. Conventional scheduling algorithms often fail to exploit the heterogeneous resources; therefore, specially designed and well-tuned scheduling algorithms are desired for achieving better quality of service. In this study, the state-of-the-art task scheduling algorithms in the cloud and fog environments are investigated in a diverse set of dimensions. Among the relevant studies published between 2018-2022 and indexed in the Web-of-Science (WOS), SCOPUS, and Google Scholar databases, eighteen studies are selected for both the cloud and fog domains from WOS and Scopus, while seventeen studies are chosen for both the cloud and fog domains from Google Scholar. Thus, a total of 106 studies are included in this survey for the detail investigation. The scheduling algorithms are broadly classified into three categories such as heuristic, meta-heuristic, and hybrid meta-heuristic followed by detailed critical analysis. It has been observed that most of the scheduling algorithms are dynamic and non-preemptive in nature, while the higher fraction of the tasks is independent in comparison to bag of tasks and workflows. Similarly, 97% of the studies focus on multiple objectives and 68% of the techniques are non-deterministic. Further, a total of twenty different scheduling objectives are identified with makespan, resource utilization, delay, load balancing, and energy consumption as the most significant metrics. The evaluation methods including simulations (51%), real experiments (4%), analytical equations (2%), and datasets (43%) etc. are surveyed. At the end, the open issues, challenges, and future directions are argued. Cloud computing, fog computing, task scheduling, workflow, Internet of Things (IoT). INDEX TERMS I. INTRODUCTION Cloud computing offers different kinds of scalable resources to end users based on pay-per-use basis by providing costThe associate editor coordinating the review of this manuscript and approving it for publication was Zhiwu Li . savings, security, flexibility, and automatic software updates etc. The users can concentrate on their personal and business objectives without bothering about the infrastructure and supporting services as the vendor is responsible for managing it. There are three service models and four deployment models provided by the cloud providers [1]. Though it is

show abstract

“…In [24], the author proposed a modified genetic algorithm in cloud computing to identify the best servers to deploy these VMs and the best VMs to use for completing tasks that have been received. The chromosome of GAS is represented by this proposed algorithm using a matrix structure that integrates the ids of jobs, VMs, and servers.…”

Section: Literature Reviewmentioning

confidence: 99%

A New Round Robin Algorithm for Task Scheduling in Real-time System

2022

IJIES

View full text Add to dashboard Cite

The modern human life is strongly reliant on real-time systems. The operating system is one of the most significant real-time systems. To meet the demands of the future, operating systems must be upgraded. Nowadays, a lot of systems are used; like mobile applications and internet of things. Task scheduling is so important in the operating system to enhance the performance in these systems. Round Robin algorithm has been widely used in task scheduling. In this paper, a Median Mean Round Robin (MMRR) algorithm is proposed to significantly enhance the performance of the Round Robin algorithm. The proposed algorithm finds an optimal dynamic time quantum └(median+ mean)/2)┘ and generated for each cluster depending on the remaining burst time of the processes. The performance has been enhanced in terms of waiting time, turnaround time and context switching. The experimental results show that the proposed algorithm outperforms ADRR, HYRR, EDRR and MARR algorithms.

show abstract

Genetic-Based Multi-objective Task Scheduling Algorithm in Cloud Computing Environment

Cited by 7 publications

References 18 publications

A Modified Genetic-Based solution for Power-Aware Placement of Virtual Machines

A Modified Genetic-Based solution for Power-Aware Placement of Virtual Machines

A Review on Task Scheduling Techniques in Cloud and Fog Computing: Taxonomy, Tools, Open Issues, Challenges, and Future Directions

A New Round Robin Algorithm for Task Scheduling in Real-time System

Contact Info

Product

Resources

About