gpugraph500 | A GPU-based Graph500 implementation

gpugraph500 is a C++ library. gpugraph500 has no bugs, it has no vulnerabilities and it has low support. You can download it from GitHub.

Recent research projects have investigated partitioning, acceleration, and data reduction techniques for improving the performance of Breadth First Search (BFS) and the related HPC benchmark, Graph500. However, few implementations have focused on cloud-based systems like Amazon's Web Services, which differ from HPC systems in several ways, most importantly in terms of network interconnect. This codebase is evaluated in a related paper, Optimizing Communication for a 2D-Partitioned Scalable BFS, presented at the 2016 High Performance Extreme Computing Conference (HPEC 2016). This implementation supports optimizations to reduce the communication overhead of an accelerated, distributed BFS on an HPC system and a smaller cloud-like system that contains GPUs. In particular, this code implements an efficient 2D partitioning scheme and allreduce implementation, as well as different CPU-based compression schemes for reducing the overall amount of data shared between nodes. Timing and Score-P profiling results detailed in the paper demonstrate a dramatic reduction in row and column frontier queue data (up to 91%) and show that compression can improve performance for a bandwidth-limited cluster.