Hershey, Pennsylvania
New York, New YorkBeijing, China
A Minimum-Aware Container Live Migration Algorithm in the Cloud Environment

A Minimum-Aware Container Live Migration Algorithm in the Cloud Environment

Peng Li (School of Computer Science and Jiangsu High Technology Research Key Laboratory for WSN, Nanjing University of Posts and Telecommunications, Nanjing, China), Huqing Nie (School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China), He Xu (School of Computer Science and Jiangsu High Technology Research Key Laboratory for WSN, Nanjing University of Posts and Telecommunications, Nanjing, China), and Lu Dong (School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing, China)
Copyright: © 2017 |Pages: 13
DOI: 10.4018/ijbdcn.2017070102
OnDemand:
(Individual Articles)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Load imbalance is a problem faced by the distributed cloud computing platform. It often requires the information collaboration by each server in the cluster to carry out the container migration. Most of the algorithms which aim to reduce the downtime do not consider migration cost of the containers and perform some unnecessary migration. In this paper, with the aim to reduce the unnecessary migration of containers, an optimal minimum migration algorithm (OMNM) is proposed. By fitting the growth rate of Docker containers in the source server, the model can estimate the growth trend of each Docker container and determine which container needs to be migrated. While ensuring the load balancing of the cluster, the number of the migration is reduced, and the utilization ratio of the resource is improved. Experimental results show that the algorithm is effective to reduce the total number of live migration of Docker containers and reduce the workload of migration. Finally, it achieves the load balancing of cloud resources.
Article Preview
Top

1. Introduction

1.1. Virtual Container Technology

The emergence of virtual container technology (e.g. Docker) has effectively solved many of the configuration problems of traditional virtual machine, such as the difficulty to maintain it. In the concept of container, each application is provided an isolated operating space, and each container contains an exclusive environment space for users. Meanwhile, if a container changes, the operating environment of other containers is not affected. Container technology is a kind of lightweight virtual technology which doesn't need hardware virtualization.

1.2. Dynamic Migration of Containers

In the actual cloud environment with Docker containers, some problems occur due to load unbalancing. For example, some servers may start too many containers, which results in the server running slowly. On the other hand, some containers may generate new data volume, which grows too fast. In order to solve the problem of load unbalancing, we must continue with the migration process of the containers. Generally, the migration of containers is divided into two types: static and dynamic. In a static migration process, the container to be migrated is immediately closed, and then a new container is opened in the destination server to complete the migration process. In this process, the whole cloud services are suspended resulting in great economic loss. Hence dynamic migration is more often used. In the dynamic migration, while the source and destination servers are determined, the information of hot containers is confirmed by the collaboration between the two servers. The basic principle of this approach is that the user hardly feels that any migration process has taken place, since the migration time does not affect services. Hence, it is opaque to users.

Complete Article List

Search this Journal:
Reset
Volume 20: 1 Issue (2025)
Volume 19: 1 Issue (2024)
Volume 18: 2 Issues (2022): 1 Released, 1 Forthcoming
Volume 17: 2 Issues (2021)
Volume 16: 2 Issues (2020)
Volume 15: 2 Issues (2019)
Volume 14: 2 Issues (2018)
Volume 13: 2 Issues (2017)
Volume 12: 2 Issues (2016)
Volume 11: 2 Issues (2015)
Volume 10: 4 Issues (2014)
Volume 9: 4 Issues (2013)
Volume 8: 4 Issues (2012)
Volume 7: 4 Issues (2011)
Volume 6: 4 Issues (2010)
Volume 5: 4 Issues (2009)
Volume 4: 4 Issues (2008)
Volume 3: 4 Issues (2007)
Volume 2: 4 Issues (2006)
Volume 1: 4 Issues (2005)
View Complete Journal Contents Listing