Nimrod/G: An Architecture of a Resource Management and Scheduling System in a Global Computational Grid

摘要

The availability of powerful microprocessors and high-speed networks ascommodity components has enabled high performance computing on distributedsystems (wide-area cluster computing). In this environment, as the resourcesare usually distributed geographically at various levels (department,enterprise, or worldwide) there is a great challenge in integrating,coordinating and presenting them as a single resource to the user; thus forminga computational grid. Another challenge comes from the distributed ownership ofresources with each resource having its own access policy, cost, and mechanism. The proposed Nimrod/G grid-enabled resource management and scheduling systembuilds on our earlier work on Nimrod and follows a modular and component-basedarchitecture enabling extensibility, portability, ease of development, andinteroperability of independently developed components. It uses the Globustoolkit services and can be easily extended to operate with any other emerginggrid middleware services. It focuses on the management and scheduling ofcomputations over dynamic resources scattered geographically across theInternet at department, enterprise, or global level with particular emphasis ondeveloping scheduling schemes based on the concept of computational economy fora real test bed, namely, the Globus testbed (GUSTO).