Scalability of Grid Simulators: An Evaluation

  • Authors:
  • Wim Depoorter;Nils Moor;Kurt Vanmechelen;Jan Broeckhove

  • Affiliations:
  • University of Antwerp, Antwerp, Belgium BE-2020;University of Antwerp, Antwerp, Belgium BE-2020;University of Antwerp, Antwerp, Belgium BE-2020;University of Antwerp, Antwerp, Belgium BE-2020

  • Venue:
  • Euro-Par '08 Proceedings of the 14th international Euro-Par conference on Parallel Processing
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Due to the distributed nature of resources in grids that cover multiple administrative domains, grid resource management cannot be optimally implemented using traditional approaches. In order to investigate new grid resource management systems, researchers utilize simulators which allows them to efficiently evaluate new algorithms on a large scale. We have developed the Grid Economics Simulator (GES) in support of research into grid resource management in general and economic grid resource management in particular. This paper compares GES to SimGrid and GridSim, two established grid simulation frameworks. We demonstrate that GES compares favourably to the other frameworks in terms of scalability, runtime performance and memory requirements. We explain how these differences are related to the simulation paradigm and the threading model used in each simulator.