The document is organized as follows:

• Requirements
• Quick start
• Supported graph formats
• Code Documentation
• Notes
• Additional Utilities
• Reporting bugs and contributing
• Publications
• cuStinger Developers
• Acknowledgements
• License

** For cuStinger Graph Algorithms please refer to **

• Nvidia Modern GPU (compute capability >= 3.0): Kerpler, Maxwell, Pascal architectures.
• CUDA toolkit v7.5 or greater. (CUDA toolkit v8.0 recommended, CUDA toolkit v7.0 supported but not tested)
• GCC or Clang host compiler with support for C++11^*. Note: the compiler must be compatible with the related CUDA toolkit version. For more information see CUDA Installation Guide
• CMake v3.5 or greater.
• 64-bit Operating System (Ubuntu tested).

^{*The next release will support C++14 (CUDA Toolkit v8.5).}

The following basic steps are required to build and execute cuStinger:

git clone https://github.com/FedericoUnivr/cuStinger
cd cuStinger/build
cmake ..
make -j
./cuStingerTest <input_graph> <options>

The syntax and the input parameters of cuStinger are explained in details in the docs/Syntax.txt file or by typing ./cuStingerTest --help.

cuStinger supports the following graph input formats:

• Market (.mtx), The University of Florida Sparse Matrix Collection
• Metis (.graph), 10th DIMACS Implementation Challenge
• SNAP (.txt), Stanford Network Analysis Project
• Dimacs9th (.gr), 9th DIMACS Implementation Challenge
• The Koblenz Network Collection (out.< name >), The Koblenz Network Collection
• Network Data Repository (.edges), Network Data Repository
• Binary (.bin)

cuStinger directly deduces the graph structure (directed/undirected) from the input file header.

cuStinger allows reading the input graph by using a fixed binary format to speed up the file loading. The binary file is generated by cuStinger with the --binary command line option.

The code documentation is located in the docs directory (doxygen html format). The documentation is also accessible online here.

• The changelog can be found here: history
• cuStinger has been written by adopting (when possible) the following coding styles:
  • LLVM Coding Standards
  • Google C++ Style Guide
• cuStinger has been checked with the following tools to ensure the code quality:
  • clang++ v4.0: warnings
  • clang-tidy: warnings and code styles
• cuStinger has been tested with the following tools: (see CodeCheck)
  • cuda-memcheck
  • valgrind v3.12
  • clang static analyzer v279

• util/count_lines.sh : some statistics about cuStinger code.
• util/uf_graph_download.sh : simple script to download and extract University of Florida Sparse Matrix Collection graphs in a directory.

If you find any bugs please report them by using the repository (github issues panel). We are also ready to engage in improving and extending the framework if you request some new features.

• Oded Green, David A. Bader, "cuStinger: Supporting dynamic graph algorithms for GPUs", IEEE High Performance Extreme Computing Conference (HPEC), 13-15 September, 2016, Waltham, MA, USA, pp. 1-6.

• Oded Green, Researcher, Georgia Institute of Technology
• Federico Busato, Ph.D. Student, University of Verona (Italy)

• Oded Green, Researcher, Georgia Institute of Technology
• Federico Busato, Ph.D. Student, University of Verona (Italy)
• James Fox, Ph.D. Student, Georgia Institute of Technology : Temporal Motif Finding
• Devavret Makkar, Ph.D. Student, Georgia Institute of Technology : Triangle Counting
• Elisabetta Bergamini, Ph.D. Student, Karlsruhe Institute of Technology (Germany) : Katz Centrality
• ...

• Grant...

BSD 3-Clause License

Copyright (c) 2017, cuStinger All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

• Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

• Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

• Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.