Energy and QoS-aware virtual machine placement approach for IaaS cloud datacenter

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 7, 2025

Abstract Virtualization technology enables cloud providers to abstract, hide, and manage the underlying physical resources of data centers in a flexible scalable manner. It allows placing multiple independent virtual machines (VMs) on single server order improve resource utilization energy efficiency. However, determining optimal VM placement is crucial as it directly impacts load balancing, consumption, performance degradation within center. Furthermore, deciding based factor usually insufficient center because many factors must be considered, ignoring them may too expensive. This paper improves new multi-objective (MVMP) algorithm using quantum particle swarm optimization (QPSO) technique. We call QPSO-MOVMP, its objective find Pareto solution for problem by balancing different goals. generates solutions that save power minimizing number running machines, avoid maintaining service level agreement (SLA), keeping loads at utilization. The experimental results show QPSO-MOVMP had superior terms consumption compared three other algorithms conventional single-objective algorithms. Simulation proposed achieves 2.4 × 10 4 watts power. outperformed others, achieving minimum 12% SLA breaches while experiencing significant surge requests from VMs. Moreover, model generated better distribution than those derived comparative method.

Language: Английский

---

Multi-Objective Optimization of Tasks Scheduling Problem for Overlapping Multiple Tower Cranes

Buildings, Journal Year: 2024, Volume and Issue: 14(4), P. 867 - 867

Published: March 22, 2024

The scheduling of tower crane operations is a complex process. Overlapping areas between cranes often lead to increased collision possibilities, resulting in additional operation complexity. Single objectives related time or economic aspects were always considered dealing with this issue, which neglected other and the relationships different objectives. Therefore, article proposes novel method for schedule prefabricated component lifting tasks on construction site, integrating multi-objective optimization model decision-making aim minimizing energy consumption costs amplitude among multiple cranes. A non-dominated sorting genetic algorithm-III (NSGA-III) written Python used as algorithm—which considers selection each order technique preference by similarity an ideal solution (TOPSIS), applied ranking Pareto front. Then, green production education integration training building project located Jinan, China case study verify that practical reasonable. results show conflicts can be effectively avoided, reduced, equipment utilization rationally distributing overlapping And top 11 solutions, priorities 1 are close same. In contrast, task 2 was assigned based balance two discovery helpful eliminate collisions, interference, frequent start stop several cranes, so realize safe, stable, efficient site.

Language: Английский

---

A Review Study on Energy Consumption in Cloud Computing

International Journal For Multidisciplinary Research, Journal Year: 2024, Volume and Issue: 6(1)

Published: Jan. 23, 2024

Cloud computing has become a fundamental technology for wide range of services, yet its increasing energy demands present substantial environmental and economic challenges. With the rapid growth services applications, number researches have been focused on saving. The need to reduce costs is constant challenge cloud providers data centers. This paper offers an extensive review issues surrounding consumption in computing, with focus algorithms associated situational awareness, consolidation, allocation, placement/migration, scheduling virtual machines containers. We conduct critical analysis studies from 2018 2023, comparing various methodologies aimed at achieving efficiency without sacrificing performance. delineates current trends, identifies gaps existing research, proposes directions future investigations. Our study emphasizes necessity cultivating sustainable practices provides valuable insights into practical implementation energy-efficient solutions environments.

Language: Английский

---

VM Placement in Cloud Computing Using Nature-Inspired Optimization Algorithms

Advances in computer and electrical engineering book series, Journal Year: 2024, Volume and Issue: unknown, P. 251 - 282

Published: Dec. 6, 2024

Cloud computing implements various techniques for the efficient utilization of resources. These resources are delivered over internet, allowing users to access and manage them remotely. Often, these provided in form virtual machines (VMs). VMs essentially software-based emulations physical computers. In a cloud data center, numerous (PMs), also called servers, host VMs. Placement Physical Machines is critical task as there many factors that need be considered. Nature-inspired optimization algorithms such Genetic Algorithms, Particle Swarm Optimization, Ant Colony Optimization inspired by natural phenomena behaviour. have been used past generate near-optimal solutions polynomial time computationally intractable problems like VM Problems (VPP). This chapter discusses how placed centers using Nature-Inspired Algorithms.

Language: Английский

---

A Multi-Objective Approach for Optimizing Virtual Machine Placement Using ILP and Tabu Search

Telecom, Journal Year: 2024, Volume and Issue: 5(4), P. 1309 - 1331

Published: Dec. 16, 2024

Efficient Virtual Machine (VM) placement is a critical challenge in optimizing resource utilization cloud data centers. This paper explores both exact and approximate methods to address this problem. We begin by presenting an solution based on Multi-Objective Integer Linear Programming (MOILP) model, which provides optimal VM Placement (VMP) strategy. Given the NP-completeness of MOILP model when handling large-scale problems, we then propose using Tabu Search (TS) algorithm. The TS algorithm designed as practical alternative for addressing these complex scenarios. A key innovation our approach simultaneous optimization three performance metrics: number accepted VMs, wastage, power consumption. To best knowledge, first application context VMP. Furthermore, metrics are jointly optimized ensure operational efficiency (OPEF) minimal expenditure (OPEX). rigorously evaluate through extensive simulation scenarios compare its results with those enabling us assess quality relative one. Additionally, benchmark against existing literature emphasize advantages. Our findings demonstrate that strikes effective balance between practicality, making it robust VMP environments. outperforms other algorithms considered simulations, achieving gain 2% 32% OPEF, worst-case increase up 6% OPEX.

Language: Английский

---

PPO-based deployment and phase control for movable intelligent reflecting surface

Journal of Cloud Computing Advances Systems and Applications, Journal Year: 2023, Volume and Issue: 12(1)

Published: Dec. 1, 2023

Abstract Intelligent reflecting surface (IRS) stands as a promising technology to revolutionize wireless communication by manipulating incident signal amplitudes and phases enhance system performance. While existing research primarily centers around optimizing the phase shifts of IRS, deployment IRS on movable platforms introduces new degree freedom in design IRS-assisted systems. Leveraging flexible strategies for holds potential further amplify network throughput extend coverage. This paper addresses challenging non-convex joint optimization problem proposes dynamic algorithm based proximal policy (PPO) dynamically aerial position configuration IRS. Simulation results show effectiveness proposed approach, demonstrating significant performance improvements compared schemes without assistance conventional static methods.

Language: Английский

---

Systematic Literature Review on Bio Inspired Algorithms in Cloud Fog Computing

Published: Dec. 14, 2023

With the dynamic nature of modern computing landscapes, cloud and fog systems have become integral in processing delivering services. In response to challenges posed by these distributed systems, bio-inspired algorithms, such as GA, PSO ACO emerged promising tools for optimizing resource allocation, load balancing, energy efficiency, various other aspects cloud-fog computing. This review provides a comprehensive overview existing research, analyzing application, strengths, limitations algorithms environments. It categorizes discusses their use across provisioning, task scheduling, fault tolerance, security, management, shedding light on adaptability potential enhance system performance. By pinpointing areas where further research is needed providing foresight into upcoming directions, this emerges valuable reference researchers, practitioners, decision-makers. furnishes fundamental comprehension how contribute enhancing efficiency performance our interconnected world.

Language: Английский

---