Decision support for virtual machine re-provisioning in production environments
LISA'07 Proceedings of the 21st conference on Large Installation System Administration Conference
Usher: an extensible framework for managing custers of virtual machines
LISA'07 Proceedings of the 21st conference on Large Installation System Administration Conference
STORM: simple tool for resource management
LISA'08 Proceedings of the 22nd conference on Large installation system administration conference
Sysman: a virtual file system for managing clusters
LISA'08 Proceedings of the 22nd conference on Large installation system administration conference
Cloud-Oriented Virtual Machine Management with MLN
CloudCom '09 Proceedings of the 1st International Conference on Cloud Computing
Suspending, migrating and resuming HPC virtual clusters
Future Generation Computer Systems
Scalable data center provisioning and control
IBM Journal of Research and Development
Using Network Node Description Language for modeling networking scenarios
Advances in Engineering Software
Simplified cloud-oriented virtual machine management with MLN
The Journal of Supercomputing
A virtual test environment for MPI development: quick answers to many small questions
PVM/MPI'07 Proceedings of the 14th European conference on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Hi-index | 0.00 |
As the number of available virtualization tools and their popularity continues to grow, the way in which virtual machines can be managed in a data center is becoming more and more important. A few commercial tools such as VMware ESX and XenEnterprise exist, but they are limited to a certain virtual machine technology and offer no way to expand the tool's capabilities to local needs. This paper discusses an open source management tool, MLN, for large virtual networks transparent of their virtualization platform. The current version supports the two popular open source virtual machine packages: Xen and User Mode Linux. MLN uses an extensible configuration language for the design of the virtual machines and their internal configuration. Large groups of virtual machines can be managed as logical groups. We present a web-server hosting scenario and a on-demand render farm as case studies to show the usefulness of our tool. The text concludes with a short discussion on the difficulties of offering abstraction to virtualization platforms.