|
      
|
|
|
|
A Grid computational environment is suitable for executing parallel applications. In spite of such suitability, few users currently execute their parallel applications on Grids. The main reason for this is the dynamically heterogeneous environment of Grids, which creates a complex problem for the design of task mapping methods.
We use the clustering and mapping method to understand and determine the impact of dynamic resource change for a parallel application running on a Grid. In this talk we will present an overview of our work and our first steps toward a Grid Scheduler Testing environment.
