A Grid Advance Reservation Framework for Co-allocation and Co-reservation Across Heterogeneous Local Resource Management Systems

Abstract

Co-allocation and co-reservation is a key capability of Grid schedulers for supporting some complex Grid applications, e.g., workflow. The chief enabling technology of co-allocation and co-reservation is Advance Reservation (AR), which is typically implemented by local Resource Management Systems (RMSs). As at present only a limited number of RMSs can support AR, and most of them use individual interface formats, it is rather difficult for Grid schedulers to manage Grid ARs across heterogeneous RMSs, including AR-incapable ones, through a uniform interface. In this paper we propose a Grid AR framework which can address the issue by means of a Grid AR manager that is able to externalize the AR functionality from local RMSs. An advanced Grid AR algorithm is implemented in the Grid AR manager, and a local AR API and Grid AR API is respectively defined to standardize the interaction between the Grid AR manager and local RMSs, as well as between high level Grid scheduler components and Grid AR manager. Based on a plugin architecture, the Grid AR framework is able to incorporate different types of local RMSs to implement Grid AR functionalities, irrespective of whether local RMSs support AR or not.