Using Metadata Snapshots for Extending Ant-based Resource Discovery Service in Inter-cooperative Grid Communities Ye Huang * , Nik Bessis † , Amos Brocco * , Pierre Kuonen ‡ , Michele Courant * and Beat Hirsbrunner * * Department of Informatics, University of Fribourg, Switzerland Email: {ye.huang, amos.brocco, michele.courant, beat.hirsbrunner}@unifr.ch † Department of Computing and Information Systems, University of Bedfordshire, UK Email: nik.bessis@beds.ac.uk ‡ Department of Information and Communication Technologies, University of Applied Sciences Western Switzerland Email: pierre.kuonen@hefr.ch Abstract—Much work is under way within the resource man- agement community on issues associated with grid scheduling upon dynamically discovered information. In this paper we tackle the problem by exploiting a bio-inspired resource discovery mech- anism, where information is provided by ant-based lightweight mobile agents traveling across a grid network and collecting data from each visited node. We start by providing the current state of the adopted grid scheduler, which is the result of an existing collaborative project named SmartGRID, and its underlying architecture constructed by ant-based mobile agents. We consider the problem of discovering resources in specific grid communities, which are bounded due to different shared community policies, such as diverse ant colonies, different resource discovery ap- proaches, or other issues. Several issues have been raised during the design and implementation of such infrastructure. A notable issue, namely how grid schedulers from various bounded grid communities could be used in a manner which would extend current SmartGRID functionality is identified. Our shared view is that by utilizing already discovered and stored grid nodes’s metadata snapshots in the first instance we can facilitate a more convenient and efficient resource discovery operation next time. With this in mind, our paper goes on describing our shared vision with regard to this extended functionality as well as discussing the new conceptual basis and its model architecture. Index Terms—SmartGRID; Intra and inter cooperative grid architecture; MaGate; Ant-based swarm intelligence; Metadata snapshots. I. I NTRODUCTION SmartGRID is a cooperative project aiming at increasing the efficiency, robustness, and reliability of heterogeneous grid computing infrastructures [1] concerning volatile and dynamic resources. The proposed grid middleware has been designed as a generic and modular framework supporting intelligent and interoperable grid resource management using swarm intelli- gence algorithms and multi-type grid scheduling. SmartGRID uses a layered architecture and aims at filling the gap between grid applications, which act as the resource consumers, and the grid resource low-level management systems, which behave as the resource providers. To achieve this goal, SmartGRID uses an autonomic and evolutional grid community composed of its grid schedulers, the MaGate scheduler [2]. Two approaches are currently available for discovering candidate nodes for a specific task. The first approach assumes that each node has partial knowledge of up to six direct neigh- bor nodes, which are maintained by the resource discovery service of the host node. When the host node requires the discovery of a remote node with the required features, in order to delegate a job, the direct neighbors list is exploited with the node negotiating job delegation with each member. In the second approach, when the host node joins an existing bounded grid community, it prepares a profile named agree- ment offer to disseminate its capabilities across the community. We assume that the aforementioned public profile is kept up- to-date through nodes lifecycle. This approach is not limited to one community only: each individual node is free to join multiple existing grid communities, thus publishing different capability profiles. Currently the discovered information for each specific task is discarded after its usage. We aim at extending this model such that each node of a SmartGRID community might also be capable of keeping a metadata snapshot of known remote nodes, in order to facilitate a more efficient and intelligent behavior towards relevant scheduling decisions. Moreover, as the SmartGRID architecture strives to provide intelligent scheduling for the scope of serving the grid community as a whole, not just for a single grid node, our extended work is also concerned with the design of a scheduling strategy supporting the combination of various interoperable bounded grid communities. In general, the mission of a grid scheduler is to discover appropriate resources for executing jobs across a grid commu- nity. Our vision is that of a wider grid community scheduling process is able to exploit resources in large and partially unknown grid communities, and dealing with continuously changing job queues. Thus, since each community node is supposed to receive jobs from both its local and remote grid communities, management of the job queue must deal with a more dynamic, fluid and unexpected environment. It is our goal to ensure robustness, reliability, efficiency, and