HornetAlg

The repository provides the graph algorithms implemented on top Hornet data structure.

For additional information concerning the data structure and its APIs please refer to Hornet repository.

Getting Started

The document is organized as follows:

• Requirements
• Quick start
• Hornet Algorithms
• Performance
• Hornet Algorithms Lines of Code
• Code Documentation
• Reporting bugs and contributing
• Publications
• Hornet Developers
• Acknowledgements
• License

Requirements

• Nvidia Modern GPU (compute capability ≥ 3.0): Kerpler, Maxwell, Pascal, Volta architectures.
• CUDA toolkit 9 or greater.
• GCC or Clang host compiler with support for C++14. Note: the compiler must be compatible with the related CUDA toolkit version. For more information see CUDA Installation Guide.
• CMake v3.6 or greater.
• 64-bit Operating System (Ubuntu 16.04 or above suggested).

Quick Start

The following basic steps are required to build and execute the Hornet algorithms:

git clone --recursive https://github.com/hornet-gt/hornetsnest
cd hornetsnest/build
cmake ..
make p

By default, the CUDA compiler nvcc uses gcc/g++ found in the current execution search path as host compiler (cc --version to get the default compiler on the actual system). To force a different host compiler for compiling C++ files (*.cpp) you need to set the following environment variables:

CC=<path_to_host_C_compiler>
CXX=<path_to_host_C++_compiler>

To force a different host compiler for compiling host side code (*.cu) substitute cmake .. with

cmake -DCUDAHC=<path_to_host_C++_compiler> ..

Note: host .cpp compiler and host side .cu compiler may be different. The host side compiler must be compatible with the current CUDA Toolkit version installed on the system (see CUDA Installation Guide).

Hornet Algorithms

Algorithm	Static	Dynamic
(BFS) Breadth-first Search	yes	on-going
(SSSP) Single-Source Shortest Path	yes	on-going
(CC) Connected Components	yes	on-going
(SCC) Strongly Connected Components	to-do	to-do
(MST) Minimum Spanning Tree	to-do	to-do
(BC) Betweeness Centrality	on-going	on-going
(PG) Page Rank	yes	on-going
(TC) Triangle Counting	yes	yes
(KC) Katz Centrality	yes	yes
(MIS) Maximal Independent Set	on-going	to-do
(MF) Maximum Flow	to-do	to-do
(CC) Clustering Coeffient	to-do	to-do
(ST) St-Connectivity	to-do	to-do
(TC) Transitive Closure	to-do	to-do
Community Detection	to-do	to-do
Temporal Motif Finding	on-going	to-do
Sparse Vector-Matrix Multiplication	yes	to-do
Jaccard indices	to-do	to-do
Energy/Parity Game	on-going	to-do

Performance

CPU vs. GPU

Algorithm	CPU1	GPU 2	Speedup
(BFS) Breadth-first Search
(SSSP) Single-Source Shortest Path
(CC) Connected Components
(MST) Minimum Spanning Tree
(BC) Betweenness Centrality
(PG) Page Rank
(TC) Triangle Counting
(KC) Katz Centrality

¹ Intel ...
² NVidia Tesla P100 ..

Static vs. Dynamic

Algorithm	Static	Dynamic	Speedup
(BFS) Breadth-first Search
(SSSP) Single-Source Shortest Path
(CC) Connected Components
(MST) Minimum Spanning Tree
(BC) Betweenness Centrality
(PG) Page Rank
(TC) Triangle Counting
(KC) Katz Centrality

Hornet Algorithms Lines of Code

Algorithm	Static (A)	Static (B)	Dynamic (A)
(BFS) Breadth-first Search	4	6
(SSSP) Single-Source Shortest Path
(CC) Connected Components
(MST) Minimum Spanning Tree
(BC) Betweenness Centrality
(PG) Page Rank
(TC) Triangle Counting
(KC) Katz Centrality

^{(A) lines of code required for the algorithm
(B) lines of code required for the operators}

Code Documentation

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

Reporting bugs and contributing

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 new features.

Publications

• 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. link
• Oded Green, James Fox, Euna Kim, Federico Busato, Nicola Bombieri, Kartik Lakhotia, Shijie Zhou, Shreyas Singapura, Hanqing Zeng, Rajgopal Kannan, Viktor Prasanna, David A. Bader, "Quickly Finding a Truss in a Haystack", IEEE/Amazon/DARPA Graph Challenge, *Innovation Awards*.
• Devavret Makkar, David A. Bader, Oded Green, Exact and Parallel Triangle Counting in Streaming Graphs, IEEE Conference on High Performance Computing, Data, and Analytics (HiPC), 18-21 December 2017, Jaipur, India, pp. 1-10.

---

If you find this software useful in academic work, please acknowledge Hornet.

---

Hornet Developers

Data Structure

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

Algorithms

• Federico Busato, Ph.D. Student, University of Verona (Italy)
• Oded Green, Researcher, Georgia Institute of Technology
• James Fox, Ph.D. Student, Georgia Institute of Technology : Maximal Independent Set, 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
• Euna Kim, Ph.D. Student, Georgia Institute of Technology : Dynamic PageRank
• ...

Acknowledgements

• Grant...

License

BSD 3-Clause License

Copyright (c) 2017, Hornet 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.