Enhanced Data Management for Virtualized Service Providers

Virtualization has been lately introduced for supporting and simplifying service providers management with promising results. Nevertheless, using virtualization introduces also new challenges that must be considered. One of them relates with the data management in the provider. In this paper, we propose an innovative approach for performing efficiently all the data-related processes in a virtualized service provider, namely VM creation, VM migration and data stage-in and stage-out. Our solution provides a global repository where clients can upload the task input files and retrieve the output files. In addition, the provider implements a distributed file system (using NFS) in which each node can access its own local disk and the disk of the other nodes. This allows each node creating VMs and executing tasks efficiently using the local disk. Furthermore, tasks can be migrated with minimum overhead, since it is not necessary to transfer the VM images, while maintaining their accessibility during the whole process. Moreover, migrated tasks can be executed accessing remotely to the disk without noticeable performance penalty as demonstrated in our evaluation.

[1]  Renato J. O. Figueiredo,et al.  VMPlants: Providing and Managing Virtual Machine Execution Environments for Grid Computing , 2004, Proceedings of the ACM/IEEE SC2004 Conference.

[2]  Ian T. Foster,et al.  Virtual Workspaces in the Grid , 2005, Euro-Par.

[3]  Xiaomin Zhu,et al.  From virtualized resources to virtual computing grids: the In-VIGO system , 2005, Future Gener. Comput. Syst..

[4]  Chris I. Dalton,et al.  SoftUDC: a software-based data center for utility computing , 2004, Computer.

[5]  Xuxian Jiang,et al.  SODA: a service-on-demand architecture for application service hosting utility platforms , 2003, High Performance Distributed Computing, 2003. Proceedings. 12th IEEE International Symposium on.

[6]  Faris Nizamic,et al.  Service-Oriented Computing , 2010, Lecture Notes in Computer Science.

[7]  Dongyan Xu,et al.  VioCluster: Virtualization for Dynamic Computational Domains , 2005, 2005 IEEE International Conference on Cluster Computing.