An experimental comparison of Grid5000 clusters and the EGEE grid
Abstract
In this paper, we present a set of experiments comparing the EGEE production infrastructure and the Grid5000 experimental one. Our goal is to better understand and quantify how these systems behave under load. We first identify specific characteristics of the workload and data management systems of these two infrastructures, underlining some of their limitations and suggesting some improvements. We experimentally determine characteristic parameters for each infrastructure. From them, we build a multi-grids model that minimizes the execution time of an application by finding the best repartition of jobs between these two systems. We finally coherently compare Grid5000 clusters and EGEE using a real medical imaging application.
Origin : Files produced by the author(s)
Loading...