For over a decade, Desktop Grid (DG) systems have paved the way to high throughput computing over large scale network of Desktops. Nowadays, the aggregate computing power of the main Volunteer Computing projects shows performance exceeding several PetaFlops.Cloud computing is Internet-based computing, whereby shared resources, software, and information are provided to computers and other devices on demand, like the electricity grid . This project mainly focuses on the interoperability between Desktop Grid and Cloud. We based our work on XtremWeb-HEP, CRANE and VAST to demonstrate DG and Cloud interoperability.

Research Goals:

• Cloud resource management: On-demand resource allocation for desktop jobs
• Interoperation Portal: Job queueing and job formalization
• Performance evaluation: porting VAST application to run on this inter-operative ecosystem
• Reliability and QoS: Improving reliability/ QoS of DG with cloud

Challenges:

• Cloud to desktop grid job delegation.

Involved research institutes:

• Service Computing Technology and Systems Lab/Cluster and Grid Computing Lab,Huazhong University of Science and Technology, China
• Laboratoire de l'Accélérateur Linéaire, France

The picture below illustrates the interoperation between XtremWeb-HEP and CRANE as well as running VAST Image Search Engine on the ecosystem.