VM2Docker : automating the conversion from virtual machine to docker container
Author(s)
Lubin, Eric, M. Eng. Massachusetts Institute of Technology
DownloadFull printable version (1.602Mb)
Alternative title
Automating the conversion from virtual machine to docker container
Other Contributors
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.
Advisor
Jim Yang and Martin C. Rinard.
Terms of use
Metadata
Show full item recordAbstract
Container technology represents a flourishing field in cloud computing. For many types of computing, containers are a viable alternative to virtual machines because many applications do not require isolated kernels. Containers share the kernel with the host, as opposed to virtual machines which have a completely isolated kernel. Because of this distinction, containers are more lightweight and higher performing, but also have less isolation and therefore increased security concerns. The Docker framework, among other alternatives, has gotten the most attention and popularity over the past year and provides a powerful layered filesystem to improve deployability and provide space savings for those containers that share many layers in common. As of this writing, there is no system for automatically converting VMs to containers, as all configuration must be done manually. This is potentially unwieldy for system administrators looking to convert five to ten, or even hundreds, of virtual machines at once. This thesis presents a system we call VM2Docker that attempts to automate this conversion. VM2Docker specifically focuses on automatically generating layers for Docker to take advantage of the filesystem similarities across VMs of the same operating system. VM2Docker has been tested on various releases of Ubuntu, CentOS, and Mageia with a large degree of success and is able to provide filesystem space savings and deployment speed improvements with as few as 2 instances of a VM of a given operating system and release.
Description
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2015. Cataloged from PDF version of thesis. Includes bibliographical references (pages 66-68).
Date issued
2015Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer SciencePublisher
Massachusetts Institute of Technology
Keywords
Electrical Engineering and Computer Science.