Energy Efficient VM Live Migration and Allocation at Cloud Data Centers

Energy Efficient VM Live Migration and Allocation at Cloud Data Centers

Djouhra Dad, Djamel Eddine Yagoubi, Ghalem Belalem
Copyright: © 2014 |Pages: 9
DOI: 10.4018/ijcac.2014100105
OnDemand:
(Individual Articles)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Aiming at data center virtual machines Migration, allocating resource dynamically in order to reduce energy is a significant problem in cloud. This energy doesn't cause only the decrease of cloud provider's profit but also emit a large amount of carbon dioxide. This paper studies the resource allocation and live migration of Virtual Machines (VMs). It proposes a Double Threshold Migration (DTM) algorithm which takes into consideration an upper and a lower threshold of CPU utilization. These Thresholds let one select a number of VMs to do the migration. The live migration of the VMs reduces the high utilization of the servers and set on off state the unused physical machines (PMs). To solve the problem of the VM placement, the work applies a modification of the Best Fit Decreasing (MBFD) algorithm. Experiment results show that the proposed approach improve resource utilization, reduce the energy consumption and maintain the SLA (Service Level Agreement) violations with the energy constraint.
Article Preview
Top

Authors in (Beloglazov et al., 2011a) have proposed architecture of a system and three high-level power management policies. The authors have focused on managing the energy consumption of processors. The DVFS technique was used. The authors have introduced two phases to optimize the allocation of VMs. The first one is the selection of the VM to be migrated. In this phase three policies have been proposed: Minimization of Migration policy (MM), Highest Potential Growth policy (HPG) and Random Choice policy (RC). The second phase is the placement of the selected VM by the proposed policies. The algorithm MBFD (Modified Best Fit Decreasing) was used to allocate the VMs. Several experiments have been implemented.

Complete Article List

Search this Journal:
Reset
Volume 14: 1 Issue (2024)
Volume 13: 1 Issue (2023)
Volume 12: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 4 Issues (2016)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing