Autonomous DRL-based energy efficient VM consolidation for cloud data centers

Physical Communication, Journal Year: 2022, Volume and Issue: 55, P. 101925 - 101925

Published: Oct. 21, 2022

Language: Английский

---

Multi-objective Prediction-Based Optimization of Power Consumption for Cloud Data Centers

Arabian Journal for Science and Engineering, Journal Year: 2022, Volume and Issue: 48(2), P. 1173 - 1191

Published: March 18, 2022

Language: Английский

---

GMPR: A Two-Phase Heuristic Algorithm for Virtual Machine Placement in Large-Scale Cloud Data Centers

IEEE Systems Journal, Journal Year: 2022, Volume and Issue: 17(1), P. 1419 - 1430

Published: July 20, 2022

In the cloud data centers, due to variable resource requirements of tenants, designers applied infrastructure as a service (IaaS) model provide services for tenants with allocating in charging. The application virtualization technology enables multiple virtual machines (VMs) share resources physical machine (PM). Meanwhile, efficiency centers greatly depends on working VMs. Virtual placement (VMP) plays vital role minimizing total energy consumption and wastage (CDCs). this article, we propose greedy algorithm power (GMPR) VMP scheme address abovementioned issues. GMPR prioritizes power-efficiency PM reduce number active PMs minimize consumption. addition, reducing involves first balance novel hosts VM second placed VM. Extensive simulation results are conducted synthetic instances Amazon EC2 performance metrics confirm that has superiority by an average 1.91% 16.18% compared cutting-edge method.

Language: Английский

---

A Novel Self-Adaptive VM Consolidation Strategy Using Dynamic Multi-Thresholds in IaaS Clouds

Future Internet, Journal Year: 2018, Volume and Issue: 10(6), P. 52 - 52

Published: June 13, 2018

With the rapid development of cloud computing, demand for infrastructure resources in data centers has further increased, which already led to enormous amounts energy costs. Virtual machine (VM) consolidation as one important techniques Infrastructure a Service clouds (IaaS) can help resolve consumption by reducing number active physical machines (PMs). However, necessity considering energy-efficiency and obligation providing high quality service (QoS) customers is trade-off, aggressive may lead performance degradation. Moreover, most existing works threshold-based VM strategy are mainly focused on single CPU utilization, although resource request different VMs very diverse. This paper proposes novel self-adaptive based dynamic multi-thresholds (DMT) PM selection, be dynamically adjusted future utilization multi-dimensional CPU, RAM Bandwidth. Besides, selection placement algorithm also improved utilizing each parameter DMT. The experiments show that our proposed better than other strategies, not only QoS but less consumption. In addition, advantage its reduction hosts much more obvious, especially when it under extreme workloads.

Language: Английский

---

Optimizing Virtual Machines Placement in a Heterogeneous Cloud Data Center System

International Journal of Computer Networks And Applications, Journal Year: 2024, Volume and Issue: 11(1), P. 1 - 1

Published: Feb. 26, 2024

In a cloud computing environment, good resource management remains major challenge for its operation.Implementing virtual machine placement (VMP) on physical machines helps to achieve various objectives, such as allocation, load balancing, energy consumption, and quality of service.VMP (virtual placement) in the is critical, so it's important audit implementation.It must take into account resources server, including CPU, RAM, storage.In this paper, metaheuristic algorithm based Grey Wolf Optimization (GWO) method used optimize effectively minimizing number active host servers.Experimental results demonstrate effectiveness proposed method, called Virtual Machine Placement (GWOVMP).The reduces power consumption by 20.99 wastage 1.80 compared with existing algorithms.

Language: Английский

---