Overview

Official web site: http://simgrid.gforge.inria.fr/

SimGrid is a scientific instrument to study the behavior of large-scale distributed systems such as Grids, Clouds, HPC or P2P systems. It can be used to evaluate heuristics, prototype applications or even assess legacy MPI applications. All this as a free software.

The SimGrid project is 10 years old, and ready for at least 10 more thanks to the support of several agencies: the ANR (french research funding agency) funded the USS-SimGrid and SONGS scientific projects; The Inria (french research institute in computer science) funds engineers and provide our infrastructure and many of us are paid to work on SimGrid.

SimGrid is Versatile

  • SimGrid provides ready to use models and API to simulate many different distributed systems: clusters, wide-area and local-area networks, peers over DSL connexions, data centers, etc.
  • SimGrid as a Grid Simulator: Accurate yet fast simulation models
  • SimGrid as a P2P Simulator: Highly scalable simulations (several milions of nodes on a single machine)
  • SimGrid as a MPI Simulator: Realistically simulates unmodified MPI programs
  • SimGrid as a Cloud Simulator: SimGrid provides an libvirt-like interface for your Cloud simulation

SimGrid is Trustfully and Usable

  • Our sound simulation models are theoretically and experimentally assessed.
  • Scalable: Fast, low memory footprint.
  • Portable: Linux / Mac OS X / Windows ; Numerous bindings (write your simulators in C, Lua or Java or Ruby)
  • Produces traces that can easily be visualized and post-processed with state-of-the-art tools such as Pajé and Viva.
  • SimGrid provides many fancy features: synthetic platforms generation, traces for the platform availability or for the applicative workload, dynamicity and churn, etc.

SimGrid is Open

  • SimGrid is free software. It is distributed under the GPL license, and all developments occure in open repositories.
  • SimGrid is widely used around the globe. It grounded the experiments of numerous scientific papers. It is used in practice at Cern to improve the algorithms basing the data management infrastructure and by users of the EGEE grid to improve their usage of the platform and adapt to its variations.

Issue tracking

View all issues