[simgrid.git] / doc / index.doc

/*! \page index

<center>
\htmlonly
<img align=center src="simgrid_logo.png" alt="SimGrid"><br>
\endhtmlonly
</center>

\section hot_news HOT NEWS

<b>SimGrid 3.0.1 was released on 22st of October 2005. Get it from
<a href="https://gforge.inria.fr/frs/?group_id=12">here</a>.</b> 

\section quick Quick start

SimGrid is a toolkit that provides core functionalities for the simulation
of distributed applications in heterogeneous distributed environments.
The specific goal of the project is to facilitate research in the area of
distributed and parallel application scheduling on distributed computing
platforms ranging from simple network of workstations to Computational
Grids.

\subsection quick_dl Getting and installing the software

  - The official webpage is <a href="http://simgrid.gforge.inria.fr/">simgrid.gforge.inria.fr</a>. 
  - The development webpage is <a href="https://gforge.inria.fr/projects/simgrid//">gforge.inria.fr/projects/simgrid</a>. 
  - The user mailing list is <simgrid-user@lists.gforge.inria.fr>
  - The SimGrid software package can be downloaded from <a href="https://gforge.inria.fr/frs/?group_id=12">here</a>. 

To compile and install it, simply type the following. If you are not
familiar with compiling C files under UNIX and using libraries, please check
the \ref faq. SimGrid also works under Windows, but we do not distribute any
pre-compiled binaries [yet].

\verbatim $ ./configure
 $ make all
 [become root]
 # make install\endverbatim
 
\subsection quick_more More information

The API (along with some example of
use) is described in \ref SimGrid_API.

For more information about the SimGrid toolkit, please simply keep reading
this page. It is organized as follow:

  - \ref overview : Presentation of the toolkit, of each of its components
    and of their interactions.
  - \ref people : Who is behind this project.
  - \ref publications : Some articles providing more details about the
    SimGrid toolkit or using and validating it.

<hr>

\section overview Overview of the toolkit components

As depicted by the following diagram, the SimGrid toolkit is basically
three-layered (click on the picture to jump to a specific component).


\htmlonly
<center>
\endhtmlonly
\htmlinclude simgrid_modules.map
\htmlonly
<br><b>Relationships between the SimGrid components</b>
</center>
\endhtmlonly

\subsection overview_fondation Base layer

The base of the whole toolkit is constituted by the <b>\ref XBT_API
(eXtended Bundle of Tools)</b>.

It is a portable library providing some grounding features such as \ref
XBT_log, \ref XBT_ex and \ref XBT_config. XBT also encompass
the following convenient datastructures: \ref XBT_dynar, \ref XBT_fifo, \ref
XBT_dict, \ref XBT_heap, \ref XBT_set and \ref XBT_swag.

See the \ref XBT_API section for more details.

\subsection overview_kernel Simulation kernel layer

The core functionnalities to simulate a virtual platform are provided by a
module called <b>\ref SURF_API</b> ("that's historical, my friend").  It is
very low-level and is not intended to be used as such by end-users. Instead,
it serve as a basis for the higher level layer.

SURF main features are a fast max-min linear solver and the ability to
change transparently the model used to describe the platform. This greatly
eases the comparison of the several models existing in the litterature.

See the \ref SURF_API section for more details.

\subsection overview_envs Programmation environments layer

This simulation kernel is used to build several programmation environments.
Each of them target a specific audiance and constitute a different paradigm.
To choose which of them you want to use, you have to think about what you
want to do and what would be the result of your work. 

 - If you want to study a theoritical problem and compare several
   heuristics, you probably want to try <b>\ref MSG_API</b> (yet another
   historical name). It was designed exactly to that extend and should allow
   you to build easily rather realistic multi-agents simulation. Yet,
   realism is not the main goal of this environment and the most annoying
   technical issues of real platforms are masked here. Check the \ref
   MSG_API section for more information.

 - If you want to study the behaviour of a MPI application using emulation
   technics, you should have a look at the <b>\ref SMPI_API</b> (Simulated
   MPI) programming environment. Unfortunately, this work is still underway.
   Check the \ref SMPI_API section for more information. 
   
 - If you want to develop a real distributed application, then you may find
   <b>\ref GRAS_API</b> (Grid Reality And Simulation) useful. This is an API
   for the realization of distributed applications. 
   \n\n
   Moreover, there is two implementations of this API: one on top of the
   SURF (allowing to develop and test your application within the comfort of
   the simulator) and another suited for deployment on real platforms
   (allowing the resulting application to be highly portable and extremely
   efficient).
   \n\n
   Even if you do not plan to run your code for real, you may want to switch
   to GRAS if you intend to use MSG in a very intensive way (e.g. for
   simulating a peer-to-peer environment).
   \n\n
   See the \ref GRAS_API section for more details.

If your favorite programming environment/model is not there (BSP,
components, etc.) is not represented in the SimGrid toolkit yet, you may
consider adding it. You should contact us first on the 
<a href=http://lists.gforge.inria.fr/mailman/listinfo/simgrid-devel>SimGrid 
developers mailing list</a>, though.

Any question, remark or suggestion are welcome on the 
<a href=http://lists.gforge.inria.fr/mailman/listinfo/simgrid-user>SimGrid users
mailing list</a>.

<hr>

\section people People

SimGrid is a joint project between University of Hawai at Manoa, ID
Laboratory (Grenoble, France) and University of Nancy (France). The
authors of SimGrid are:

 - Henri Casanova <casanova#cs.ucsd.edu> (Information and Computer Sciences Department, University of Hawai`i at Manoa)
 - Arnaud Legrand <arnaud.legrand#imag.fr> ("Informatique et Distribution" Laboratory (Grenoble, France), MESCal project)
 - Martin Quinson <martin.quinson#loria.fr> (University of Nancy (Nancy, France), LORIA Laboratory, Algorille project)

\subsection contributers Contributers and alumni project members

 - Loris Marchal: wrote the algorithm for simulation TCP bandwidth-sharing.
 - Julien Lerouge : wrote a XML parser for ENV descriptions and helped for
   the general design during a 4 month period (march-june 2002) 
   in the LIP.
 - Clément Menier and Marc Perache : wrote a first prototype of the MSG
   interface during a project at ENS-Lyon (jan 2002). 
 - Dmitrii Zagorodnov : wrote some parts of the first version of SimGrid
   (1999). 

<hr>

\section publications Selected publications

When citing SimGrid, the prefered reference paper is <i>Scheduling
Distributed Applications: the SimGrid Simulation Framework</i>, even if it's
a bit old now. We are actively working on improving this.

\subsection simulation About simulation

\li <b>Scheduling Distributed Applications: the
       SimGrid Simulation Framework</b>\n
    by <em>Henri Casanova and Arnaud Legrand and Loris Marchal</em>\n
    Proceedings of the third IEEE International Symposium
    on Cluster Computing and the Grid (CCGrid'03)\n
    Since the advent of distributed computer systems an active field
    of research has been the investigation of scheduling strategies
    for parallel applications.  The common approach is to employ
    scheduling heuristics that approximate an optimal
    schedule. Unfortunately, it is often impossible to obtain
    analytical results to compare the efficacy of these heuristics.
    One possibility is to conducts large numbers of back-to-back
    experiments on real platforms.  While this is possible on
    tightly-coupled platforms, it is infeasible on modern distributed
    platforms (i.e. Grids) as it is labor-intensive and does not
    enable repeatable results. The solution is to resort to
    simulations. Simulations not only enables repeatable results but
    also make it possible to explore wide ranges of platform and
    application scenarios.\n
    In this paper we present the SimGrid framework which enables the
    simulation of distributed applications in distributed computing
    environments for the specific purpose of developing and evaluating
    scheduling algorithms.  This paper focuses on SimGrid v2, which
    greatly improves on the first version of the software with more
    realistic network models and topologies.  SimGrid v2 also enables
    the simulation of distributed scheduling agents, which has become
    critical for current scheduling research in large-scale platforms.
    After describing and validating these features, we present a case
    study by which we demonstrate the usefulness of SimGrid for
    conducting scheduling research.


\li <b>A Network Model for Simulation of Grid Application</b>\n
    by <em>Henri Casanova and Loris Marchal</em>\n
    \anchor paper_tcp
    In this work we investigate network models that can be
    potentially employed in the simulation of scheduling algorithms for
    distributed computing applications. We seek to develop a model of TCP
    communication which is both high-level and realistic. Previous research
    works show that accurate and global modeling of wide-area networks, such
    as the Internet, faces a number of challenging issues. However, some
    global models of fairness and bandwidth-sharing exist, and can be link
    withthe behavior of TCP. Using both previous results and simulation (with
    NS), we attempt to understand the macroscopic behavior of
    TCP communications. We then propose a global model of the network for the
    Grid platform. We perform partial validation of this model in
    simulation. The model leads to an algorithm for computing
    bandwidth-sharing. This algorithm can then be implemented as part of Grid
    application simulations. We provide such an implementation for the
    SimGrid simulation toolkit.\n
    ftp://ftp.ens-lyon.fr/pub/LIP/Rapports/RR/RR2002/RR2002-40.ps.gz


\li <b>MetaSimGrid : Towards realistic scheduling simulation of
        distributed applications</b>\n
    by <em>Arnaud Legrand and Julien Lerouge</em>\n
    Most scheduling problems are already hard on homogeneous
    platforms, they become quite intractable in an heterogeneous
    framework such as a metacomputing grid. In the best cases, a
    guaranteed heuristic can be found, but most of the time, it is
    not possible. Real experiments or simulations are often
    involved to test or to compare heuristics. However, on a
    distributed heterogeneous platform, such experiments are
    technically difficult to drive, because of the genuine
    instability of the platform. It is almost impossible to
    guarantee that a platform which is not dedicated to the
    experiment, will remain exactly the same between two tests,
    thereby forbidding any meaningful comparison. Simulations are
    then used to replace real experiments, so as to ensure the
    reproducibility of measured data. A key issue is the
    possibility to run the simulations against a realistic
    environment. The main idea of trace-based simulation is to
    record the platform parameters today, and to simulate the
    algorithms tomorrow, against the recorded data: even though it
    is not the current load of the platform, it is realistic,
    because it represents a fair summary of what happened
    previously. A good example of a trace-based simulation tool is
    SimGrid, a toolkit providing a set of core abstractions and
    functionalities that can be used to easily build simulators for
    specific application domains and/or computing environment
    topologies. Nevertheless, SimGrid lacks a number of convenient
    features to craft simulations of a distributed application
    where scheduling decisions are not taken by a single
    process. Furthermore, modeling a complex platform by hand is
    fastidious for a few hosts and is almost impossible for a real
    grid. This report is a survey on simulation for scheduling
    evaluation purposes and present MetaSimGrid, a simulator built
    on top of SimGrid.\n
    ftp://ftp.ens-lyon.fr/pub/LIP/Rapports/RR/RR2002/RR2002-28.ps.gz

\li <b>SimGrid: A Toolkit for the Simulation of Application
        Scheduling</b>\n
    by <em>Henri Casanova</em>\n
    Advances in hardware and software technologies have made it
    possible to deploy parallel applications over increasingly large
    sets of distributed resources. Consequently, the study of
    scheduling algorithms for such applications has been an active area
    of research. Given the nature of most scheduling problems one must
    resort to simulation to effectively evaluate and compare their
    efficacy over a wide range of scenarios. It has thus become
    necessary to simulate those algorithms for increasingly complex
    distributed, dynamic, heterogeneous environments. In this paper we
    present SimGrid, a simulation toolkit for the study of scheduling
    algorithms for distributed application. This paper gives the main
    concepts and models behind SimGrid, describes its API and
    highlights current implementation issues. We also give some
    experimental results and describe work that builds on SimGrid's
    functionalities.\n
    http://grail.sdsc.edu/papers/simgrid_ccgrid01.ps.gz

\subsection research Papers using SimGrid results

\li <b> A study of meta-scheduling architectures for high throughput
        computing: Pull vs. Push</b>\n
    by <em> Vincent Garonne, Andrei Tsaregorodtsev, and Eddy Caron </em>\n
    Proceedings of 4th Internationnal Symposium on Parallel and
    Distributed Computing Job Scheduling Strategies for Parallel
    Processing (ISPDC'05), July 2005.\n
    Preliminary version in http://marwww.in2p3.fr/~garonne/garonne-meta.pdf

\li <b>Exploiting Replication and Data Reuse to Efficiently Schedule
       Data-intensive Applications on Grids</b>\n
    by <em> E. Santos-Neto, W. Cirne, F. Brasileiro, A. Lima.</em>\n
    Proceedings of 10th Job Scheduling Strategies for Parallel Processing, June 2004.\n
    http://www.lsd.ufcg.edu.br/~elizeu/articles/jsspp.v6.pdf

\li <b>Optimal algorithms for scheduling divisible workloads on 
       heterogeneous systems</b>\n
    by <em>Olivier Beaumont and Arnaud Legrand and Yves Robert</em>\n
    in Proceedings of the 17th International Parallel and Distributed Processing Symposium (IPDPS'03).\n
    Preliminary version on ftp://ftp.ens-lyon.fr/pub/LIP/Rapports/RR/RR2002/RR2002-36.ps.gz

\li <b>On-line Parallel Tomography</b>\n
    by <em>Shava Smallen</em>\n
    Masters Thesis, UCSD, May 2001
\li <b>Applying Scheduling and Tuning to On-line Parallel Tomography </b>\n
     by <em>Shava Smallen, Henri Casanova, Francine Berman</em>\n
     in Proceedings of Supercomputing 2001
\li <b>Heuristics for Scheduling Parameter Sweep applications in
         Grid environments</b>\n
    by <em>Henri Casanova, Arnaud Legrand, Dmitrii Zagorodnov and 
            Francine Berman</em>\n
    in Proceedings of the 9th Heterogeneous Computing workshop 
    (HCW'2000), pp349-363.
*/