A major challenge in a distributed, inherently dynamic Grid is fault tolerance. The more resources and components involved, the more complicated and error-prone becomes the system. In a Grid with potentially thousands of machines connected to each other the reliability of individual resources cannot be guaranteed. This talk discusses how the fault tolerance of long-running Grid applications can be ensured. Migol is a Grid middleware, which supports the fault tolerance of Grid applications. A key feature of Migol is the ability to transparently migrate parallel applications in the Grid. Migol comprises of different services for resource allocation, selection, and application and resource monitoring. The framework is based on open standards and is built on top of the Globus Toolkit 4. In addition, this talk will discuss methods to ensure the fault tolerance of critical infrastructure services. For example, Migol replicates critical services, such as the central information service and the monitoring services, using a ring-based replication protocol to achieve data consistency.
http://www.cct.lsu.edu/events/talks/314
Andre Luckow, University of Potsdam, Germany
February 07 2008 3:00 pm
Johnston Hall Room 338 CCT
No comments:
Post a Comment