-
Notifications
You must be signed in to change notification settings - Fork 202
Windows
We have made no effort to support gunrock/gunrock natively on Windows system. However, starting gunrock/essentials (new version of gunrock), there's been initial ground work put into getting gunrock working natively on Windows.
Just fyi, if you have Windows 10 or above, you can download and install Windows Subsystem for Linux (WSL) and seamlessly run gunrock with full CUDA support.
We also recommend looking at Windows Terminal for development. If any of these links are broken or out-dated, please let us know (and also Google's your friend to find the relevant installation docs.)
Using cmake, and visual studio, gunrock/gunrock will build like just any other CUDA project. Note as of right now, you must delete the cub symbolic link within downloaded thrust directory for the build to function as intended. Please see our GitHub Action workflow for windows for reference.
rm ${{github.workspace}}/externals/thrust-src/cub- Overview
- Publications
- Presentations
- Copyright and License
- Developers and Contributors
- Frequently Asked Questions
- Acknowledgments
- Programming Model
- Gunrock Operators
- Graph Algorithms
- Getting Essentials
- Gunrock For Linux
- Gunrock For Windows
- Gunrock For Max OSX
- (GitHub Template)
gunrockproject example
- MGPU, Python, Docs (needs review)
- Boolmap Frontier
- Hypergraphs (private)
- Modern CPP Features
- Programming Interface Examples (API)
- Style Guide
- Understanding the code structure
- Git Workflow
-
Debugging with
cuda-memcheckandcuda-gdb - Profiling with NVIDIA Nsight Systems and Compute
- Unit testing with GoogleTest
- Performance analysis
- How to write a new graph algorithm
-
PageRank: From
networkxtogunrock - How to write parallel operators
- How to add a new graph representation
- How to add a new frontier representation
- How to add multiple GPU support
- How to bind an application to python
- How to use
thrust/cub