minissf is a streamlined implementation of the Scalable Simulation Framework (SSF). It is a simulator designed for parallel discrete-event simulation of large-scale complex systems. minissf can run on most high-performance computing platforms, including shared-memory multi-processor multi-core machines or distributed machines.

Follow the following steps to download, build and run minissf.

1. Configure minissf:

% cd minissf % ./configure

If you need to change the compiler flags, for example, enable the debugging information for debuggers, like gdb, and turn on all compiler warnings when you compile minissf, you can configure the code with the following flags:

% ./configure CXXFLAGS="-O0 -g -Wall" CFLAGS="-O0 -g -Wall"

You can explicitly set the MPI compilers. MPI is required by minissf to run on distributed memory platforms. If it's not set, the configure script will try to find their names from the usual suspects, such as mpicxx, mpiCC, etc. If your MPI program has an unusual name or at a different location (such as the following example), one should specify the MPI compilers explicitly:

% ./configure --with-mpicxx="/opt/usr/bin/mpicxx" --with-mpicc="/opt/usr/bin/mpicc"

minissf uses LLVM/Clang for C++ source code translation. You should be able to download and compile the latest LLVM/Clang source code tree. Note that building LLVM/Clang may take a long time, possibly well over half an hour depending on the speed of your machine. Detailed instructions on how to install LLVM and Clang can be found at http://llvm.org/. You can get the lastest source code release of LLVM/Clang from http://llvm.org/releases/download.html.

In the following example, we use LLVM version 2.9. Follow the commands below to install LLVM and Clang (which will be put in the llvm-2.9 directory under the current working directory):

% tar xvzf llvm-2.9.tgz % tar xvzf clang-2.9.tgz % mv clang-2.9 ./llvm-2.9/tools/clang % cd llvm-2.9 % ./configure % make

After that, when you configure minissf, you need to let minissf know where LLVM/Clang source code is located:

% cd minissf % ./configure --with-llvm-path=<LLVM_INSTALL_DIR>

2. Build minissf:

After minissf has been properly configured, you can build it simply using:

% make

This step will create the minissf library (libssf.a) that you would need to run minissf applications.

At any time, you can restore a clean minissf distribution (say, in order to reconfigure minissf), using:

% make clean

or,

% make distclean

The latter not only deletes all the library and object files that are created during compilation, but also remove all files related to the configuration. It's restores the entire minissf directory to the factory state (as if you just check out from svn or untar from the archive file).

3. Generate documents:

If you want to create the user's manual and reference manual, you need to do this manually:

% cd doc % make

The user's manual will be generated using Sphinx (http://sphinx.pocoo.org/), which is a python documentation generator that uses a markup language, called reStructuredText (http://docutils.sourceforge.net/rst.html). This means you need to have sphinx properly installed for minissf to generate the user's manual for you.

The reference manual has two flavors: one is for the modeler, who uses minissf to write simulation models, and the other is for minissf developer, who wants to know how minissf is implemented. Both reference manuals require doxygen, which is a source code documentation generator for C++ and Java (http://www.stack.nl/~dimitri/doxygen/index.html). This means you need to have doxygen installed for minissf to generate the reference manuals.

These manuals can be viewed online at http://www.primessf.net/minissf/.

4. Run tests and examples:

There are examples which you can use to test minissf. There are all located under the example subdirectory under minissf root directory. Please refer to the specific README files for detailed description of the examples. To compile all examples, you simply do the following:

% cd examples % make

Good luck!

Copyright (c) 2011-2014 Florida International University.

Permission is hereby granted, free of charge, to any individual or institution obtaining a copy of this software and associated documentation files (the "software"), to use, copy, modify, and distribute without restriction.

The software is provided "as is", without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and noninfringement. In no event shall Florida International University be liable for any claim, damages or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with the software or the use or other dealings in the software.

This software is developed and maintained by

Jason Liu Modeling and Networking Systems Research Group School of Computing and Information Sciences Florida International University Miami, Florida 33199, USA

You can find our research at http://www.primessf.net/.