2006/11/27 by Y. Caniou, E. Caron, Caniou, Y. +10
Computer Science · Decision Sciences · Engineering · Physics and Astronomy · #Astronomical Observations and Instrumentation #Astronomy and Astrophysical Research #Astrophysics (astro-ph) #Astrophysics and Cosmic Phenomena #Cloud Computing and Resource Management #Cosmology and Nongalactic Astrophysics (astro-ph.CO) #Distributed and Parallel Computing Systems #FOS: Physical sciences #Simulation Techniques and Applications #astro-ph #astro-ph.CO
paper · pdf · doi:10.48550/arxiv.astro-ph/0611831
submitted Nov 2006
arxiv created 2006/11/28 · openalex publication_date 2006/11/28 · arxiv updated 2015/01/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
One way to access the aggregated power of a collection of heterogeneous machines is to use a grid middleware, such as DIET, GridSolve or NINF. It addresses the problem of monitoring the resources, of handling the submissions of jobs and as an example the inherent transfer of input and output data, in place of the user. In this paper we present how to run cosmological simulations using the RAMSES application along with the DIET middleware. We will describe how to write the corresponding DIET client and server. The remainder of the paper is organized as follows: Section 2 presents the DIET middleware. Section 3 describes the RAMSES cosmological software and simulations, and how to interface it with DIET. We show how to write a client and a server in Section 4. Finally, Section 5 presents the experiments realized on Grid'5000, the French Research Grid, and we conclude in Section 6.