Colloquium Details
Scheduling for heterogeneous clusters and Grid environments
| Authors: | Thomas Rauber Universitat Bayreuth, Germany Gudula Ruenger Technische Universitat Chemnitz, Germany |
|---|---|
| Date: | March 22, 2007 |
| Time: | 15:30 |
| Location: | 220 Deschutes |
Abstract
Task-based executions have been shown to be successful for homogeneous parallel systems for many scientific applications providing a suitable degree of task parallelism. In this talk, a task-based programming model for heterogeneous systems and Grid environments is discussed. These environments need a dynamic scheduling because of a dynamically changing load and availability of the computing resources. Thus, a suitable representation of the execution activities is required where tasks can be shipped to remote computing resources for execution. For the dynamic distribution of tasks, we introduce the runtime system TGrid which is built on top of a communication-aware communication layer. The TGrid scheduler assigns tasks to computing resources taking into account spatial locality properties.
Biography
Thomas Rauber is a professor at the University Bayreuth where he holds the chair for parallel and distributed systems. His research interests include runtime libraries and programming environments for task-based executions of parallel and distributed algorithms, performance prediction and locality optimizations.
Gudula Ruenger is a professor at Chemnitz University of Technology. Her research interests include parallel scientific computing, software tools for parallel and distributed programming, scheduling algorithms and distributed workflow execution.