Describe the scientific/technical community and the scientific/technical activity using (planning to use) the EGEE infrastructure. A high-level description is needed (neither a detailed specialist report nor a list of references).
Future Grid Networks should be able to provide Quality of Service (QoS) guarantees, in order to support real world commercial applications and complex scientific simulations and computations. In this work we present a framework that provides hard delay guarantees to its Guaranteed Service (GS) users and examine its applicability to the EGEE infrastructure.
Report on the experience (or the proposed activity). It would be very important to mention key services which are essential for the success of your activity on the EGEE infrastructure.
We believe that the proposed QoS framework can be incorporated in the EGEE environment. The GS users can in fact be Virtual Organizations (VO), single applications or just different User Interface (UI) machines, using the EGEE's infrastructure. Furthermore the resources, Computing Elements (CE) capable of serving the GS users should be defined during their installation and configuration by the site‘s administrator. These CE will publish to the Information Service of EGEE (Berkely Database Information Index - BDII) not only data regarding their current load but also information needed for the operation of the proposed QoS framework. When a user wishes hard delay guarantees (GS user) for the scheduling of his tasks, then he can ask such a service from the Workload Management System (WMS). The WMS will be responsible for the registration of the user to GS resources and for the scheduling of his tasks to registered resources, capable of executing them before their deadline expires.
Describe the added value of the Grid for the scientific/technical activity you (plan to) do on the Grid. This should include the scale of the activity and of the potential user community and the relevance for other scientific or business applications
In the proposed framework no quantitative resource reservation is performed. Instead, the users and the resources simply agree upon the task load the former will generate and the latter will serve. Specifically the GS users are leaky bucket constrained, and so they follow a (ρ, σ) constrained task generation pattern, which is agreed separately with each resource during a registration phase. So using the proposed framework a GS user can choose a resource that will execute his task before its deadline expires with absolute certainty. Also based on this framework various resources types can be defined, based on whether they serve GS users or BE (best effort) users, or both.