A dynamic fog service provisioning approach for IoT applications DOI
Mohammad Faraji‐Mehmandar, Sam Jabbehdari, Hamid Haj Seyyed Javadi

et al.

International Journal of Communication Systems, Journal Year: 2020, Volume and Issue: 33(14)

Published: July 15, 2020

Summary Internet of Things (IoT) is an ecosystem that can improve the life quality humans through smart services, thereby facilitating everyday tasks. Connecting to cloud and utilizing its services are now public common, experts seek find some ways complete computing use it in IoT, which next decades will make everything online. Fog computing, where expands edge network, one way achieve objectives delay reduction, immediate processing, network congestion. Since IoT devices produce variations workloads over time, application experience traffic trace fluctuations. So knowing about distribution future required handle workload while meeting QoS constraint. As a result, context fog main objective resource management dynamic provisioning such avoids excess or dearth provisioning. In present work, we first propose distributed framework for autonomic computing. Then, provide customized version system based on control MAPE‐k loop. The makes reinforcement learning technique as decision maker planning phase support vector regression analysis phase. At end, conduct family simulation‐based experiments assess performance our introduced system. average delay, cost, violation decreased by 1.95%, 11%, 5.1%, respectively, compared with existing solutions.

Language: Английский

A survey on the computation offloading approaches in mobile edge computing: A machine learning-based perspective DOI

Ali Shakarami,

Mostafa Ghobaei‐Arani, Ali Shahidinejad

et al.

Computer Networks, Journal Year: 2020, Volume and Issue: 182, P. 107496 - 107496

Published: Aug. 22, 2020

Language: Английский

Citations

268
Metaheuristic algorithms for PID controller parameters tuning: review, approaches and open problems DOI Creative Commons
Joseph Stephen Bassi, Emmanuel Gbenga Dada, Afeez Abidemi

et al.

Heliyon, Journal Year: 2022, Volume and Issue: 8(5), P. e09399 - e09399

Published: May 1, 2022

The simplicity, transparency, reliability, high efficiency and robust nature of PID controllers are some the reasons for their popularity acceptance control in process industries around world today. Tuning parameters has been a field active research still is. primary objectives to achieve minimal overshoot steady state response lesser settling time. With exception two popular conventional tuning strategies (Ziegler Nichols closed loop oscillation Cohen-Coon's reaction curve) several other methods have employed tuning. This work accords thorough review state-of-the-art classical controller using metaheuristic algorithms. Methods appraised categorized into optimization purposes. Details algorithms, application, equations implementation flowcharts/algorithms presented. Some open problems future also major goal this is proffer comprehensive reference source researchers scholars working on controllers.

Language: Английский

Citations

229
A Survey on the Computation Offloading Approaches in Mobile Edge/Cloud Computing Environment: A Stochastic-based Perspective DOI

Ali Shakarami,

Mostafa Ghobaei‐Arani, Mohammad Masdari

et al.

Journal of Grid Computing, Journal Year: 2020, Volume and Issue: 18(4), P. 639 - 671

Published: Aug. 9, 2020

Language: Английский

Citations

146
Offloading Mechanisms Based on Reinforcement Learning and Deep Learning Algorithms in the Fog Computing Environment DOI Creative Commons
Dezheen H. Abdulazeez, Shavan Askar

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 12555 - 12586

Published: Jan. 1, 2023

Fog computing has emerged as a paradigm for resource-restricted Internet of things (IoT) devices to support time-sensitive and computationally intensive applications. Offloading can be utilized transfer resource-intensive tasks from resource-limited end resource-rich fog or cloud layer reduce end-to-end latency enhance the performance system. However, this advantage is still challenging achieve in systems with high request rate because it leads long queues nodes reveals inefficiencies terms delays. In regard, reinforcement learning (RL) well-known method addressing such decision-making issues. large-scale wireless networks, both action state spaces are complex extremely extensive. Consequently, techniques may not able identify an efficient strategy within acceptable time frame. Hence, deep (DRL) was developed integrate RL (DL) address problem. This paper presents systematic analysis using DRL algorithms offloading-related issues computing. First, taxonomy offloading mechanisms based on divided into three major categories: value-based, policy-based, hybrid-based algorithms. These categories were then compared important features, including problem formulation, techniques, metrics, evaluation tools, case studies, their strengths drawbacks, directions, mode, SDN-based architecture, decisions. Finally, future research directions open discussed thoroughly.

Language: Английский

Citations

43
PAPSO: A Power-Aware VM Placement Technique Based on Particle Swarm Optimization DOI Creative Commons
Abdelhameed Ibrahim‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬, Mostafa Noshy, Hesham Ali

et al.

IEEE Access, Journal Year: 2020, Volume and Issue: 8, P. 81747 - 81764

Published: Jan. 1, 2020

With the widespread usage of cloud computing to benefit from its services, service providers have invested in constructing large scale data centers. Consequently, a tremendous increase energy consumption has arisen conjunction with results, including remarkable rise costs operating and cooling servers. Besides, increasing significant impact on environment due emissions carbon dioxide. Dynamic consolidation Virtual Machines (VMs) into minimal number Physical (PMs) is considered as one magic solutions manage power consumption. The virtual machine placement problem critical issue for good VM consolidation. This paper proposes Power-Aware technique depending Particle Swarm Optimization (PAPSO) determine near-optimal migrated VMs. A discrete version (PSO) adopted based decimal encoding map VMs best appropriate PMs. Furthermore, an effective minimization fitness function employed reduce without violating Service Level Agreement (SLA). Specifically, PAPSO consolidates minimum PMs major constraint decrease overloaded hosts much possible. Therefore, migrations can be reduced drastically by taking consideration main sources migrations; underloaded ones. implemented CloudSim experimental results random workloads different sizes show that does not violate SLA outperforms Best Fit Decreasing algorithm (PABFD). It about 8.01%, 39.65%, 66.33%, 11.87% average terms consumed energy, migrations, host shutdowns combined metric Energy Violation (ESV), respectively.

Language: Английский

Citations

109
A Secure and Multiobjective Virtual Machine Placement Framework for Cloud Data Center DOI
Deepika Saxena, Ishu Gupta, Jitendra Kumar

et al.

IEEE Systems Journal, Journal Year: 2021, Volume and Issue: 16(2), P. 3163 - 3174

Published: July 20, 2021

To facilitate cost-effective and elastic computing benefits to the cloud users, energy-efficient secure allocation of virtual machines (VMs) plays a significant role at data centre. The inefficient VM Placement (VMP) sharing common physical among multiple users leads resource wastage, excessive power consumption, increased inter-communication cost security breaches. address aforementioned challenges, novel multi-objective machine placement (SM-VMP) framework is proposed with an efficient migration. ensures distribution resources VMs that emphasizes timely execution user application by reducing delay. VMP carried out applying Whale Optimization Genetic Algorithm (WOGA), inspired whale evolutionary optimization non-dominated sorting based genetic algorithms. performance evaluation for static dynamic comparison recent state-of-the-arts observed notable reduction in shared servers, cost, consumption time up 28.81%, 25.7%, 35.9% 82.21%, respectively utilization 30.21%.

Language: Английский

Citations

99
A proactive autoscaling and energy-efficient VM allocation framework using online multi-resource neural network for cloud data center DOI
Deepika Saxena, Ashutosh Kumar Singh

Neurocomputing, Journal Year: 2020, Volume and Issue: 426, P. 248 - 264

Published: Oct. 28, 2020

Language: Английский

Citations

90
Machine learning methods for service placement: a systematic review DOI Creative Commons
Parviz Keshavarz Haddadha, Mohammad Hossein Rezvani, Mahdi Mollamotalebi

et al.

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 57(3)

Published: Feb. 17, 2024

Abstract With the growth of real-time and latency-sensitive applications in Internet Everything (IoE), service placement cannot rely on cloud computing alone. In response to this need, several paradigms, such as Mobile Edge Computing (MEC), Ultra-dense (UDEC), Fog (FC), have emerged. These paradigms aim bring resources closer end user, reducing delay wasted backhaul bandwidth. One major challenges these new is limitation edge dependencies between different parts. Some solutions, microservice architecture, allow parts an application be processed simultaneously. However, due ever-increasing number devices incoming tasks, problem solved today by relying rule-based deterministic solutions. a dynamic complex environment, many factors can influence solution. Optimization Machine Learning (ML) are two well-known tools that been used most for placement. Both methods typically use cost function. usually way define difference predicted actual value, while ML aims minimize simpler terms, gap prediction reality based historical data. Instead explicit rules, uses Due NP-hard nature problem, classical optimization not sufficient. Instead, metaheuristic heuristic widely used. addition, ever-changing big data IoE environments requires specific methods. systematic review, we present taxonomy problem. Our findings show 96% distributed architecture. Also, 51% studies on-demand resource estimation 81% multi-objective. This article also outlines open questions future research trends. literature review shows one important trends reinforcement learning, with 56% share research.

Language: Английский

Citations

14
Utilization-aware energy-efficient virtual machine placement in cloud networks using NSGA-III meta-heuristic approach DOI

Elnaz Parvizi,

Mohammad Hossein Rezvani

Cluster Computing, Journal Year: 2020, Volume and Issue: 23(4), P. 2945 - 2967

Published: Feb. 12, 2020

Language: Английский

Citations

62
Bio-inspired virtual machine placement schemes in cloud computing environment: taxonomy, review, and future research directions DOI
Mohammad Masdari,

Sasan Gharehpasha,

Mostafa Ghobaei‐Arani

et al.

Cluster Computing, Journal Year: 2019, Volume and Issue: 23(4), P. 2533 - 2563

Published: Dec. 11, 2019

Language: Английский

Citations

60