LLview: User-level Monitoring in Computational Grids and e-Science Infrastructures

Large-scale scientic research often relies on the collaborative use of Grid and e-Science infrastructures that oer a wide variety of Grid resources for scientists. While many production Grid projects and eScience infrastructures have begun to oer services for the usage of computational resources to end-users during the past several years, the absence of a widely accepted standard for tracing resource usage of Grid users has lead to dieren t technologies among the infrastructures. Recently, the Open Grid Forum developed a set of emerging standard specications, namely the Usage Record Format (URF) and the Resource Usage Service (RUS) that aim to manage and expose user tracings. In this paper, we present the integration of these standards into the UNICORE Grid middleware that lays the foundation for valuable tools in the area of accounting and monitoring. We present the development of Grid extensions for the LLview application, which allows to monitor the utilization (e.g. usage of cluster nodes per users) of Grid resources controlled by Grid middleware systems such as UNICORE.