nvgraph.sh
CLI for nvGraph, which is a GPU-based graph analytics library written by NVIDIA, using CUDA.
Science Score: 54.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
✓Academic publication links
Links to: zenodo.org -
○Committers with academic emails
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (13.9%) to scientific vocabulary
Keywords
Repository
CLI for nvGraph, which is a GPU-based graph analytics library written by NVIDIA, using CUDA.
Basic Info
- Host: GitHub
- Owner: nodef
- License: mit
- Language: C++
- Default Branch: main
- Homepage: https://www.npmjs.com/package/nvgraph.sh
- Size: 144 KB
Statistics
- Stars: 3
- Watchers: 2
- Forks: 1
- Open Issues: 0
- Releases: 8
Topics
Metadata Files
README.md
CLI for nvGraph, which is a GPU-based graph analytics library written by NVIDIA, using CUDA.
🐚 Shell,
📜 Files,
📘 Wiki.
This is for running nvGraph functions right from the CLI with graphs in
MatrixMarket format (.mtx) directly. It just needs a x86_64 linux machine
with NVIDIA GPU drivers installed. Execution time, along with the results can be
saved in JSON/YAML file. The executable code is written in C++. You can
install this with npm install -g nvgraph.sh.
nvGraph, as mentioned above, is a GPU-based graph analytics library written by NVIDIA. It provides four core graph algorithms: Single Source Shortest Path (SSSP), PageRank, Triangle count, and Breadth First Search (BFS) traversal. Data is loaded into the GPU in Compressed Sparse Row (CSR) format, upon which computation is performed. Here, we load the graph stored in MatrixMarket format (a text-based file format for sparse matrices) into a dynamic graph data structure in the host (CPU) memory, which is then converted to CSR format and then transferred to the device (GPU) memory. Computed results are then copied back to the host memory, and written to the output file in suitable format (JSON/YAML). Note that measured time only includes the time required for computation on the GPU.
The SSSP algorithm accepts the source vertex as an argument, and returns
the shortest distance to each vertex form the source vertex. The PageRank
algorithm accepts the damping factor, tolerance, and max. iterations as
arguments, and returns the rank of each vertex. In addition to ranks of
vertices in case of the PageRank algorithm, we like to also obtain additional
analytics of the rank values, i.e., Lorenz curve, and Gini coefficient. The
Triangle count algorithm, unsurprisingly, counts the number of triangles
in the graph. The BFS traversal algorithm accepts the source vertex as an
argument, traverses the graph in breadthwise manner, and returns the distance
and predecessor of each vertex from the source vertex. Note that
detailed results are written to the output file only when the full
(-f) flag is provided.
Alas, nvGraph is now no more actively developed. NVIDIA started developing cuGraph a collection of graph analytics that process data found in GPU Dataframes as part of RAPIDS.
Stability: Experimental.
```bash
Finds single source shortest path from source vertex
→ returns distances
$ nvgraph sssp -o=out.json -f web-Google.mtx -s=1
Finds pagerank of all vertices
→ returns ranks
$ nvgraph pagerank -o=out.json -f web-Google.mtx -a=0.85 -t=1e-6
Counts triangles in undirected, lower triangular graph
→ returns count
$ nvgraph triangle-count -o=out.json -f web-Google.mtx
Traverses breadth-first from source vertex
→ returns distances, predecessors
$ nvgraph traversal-bfs -o=out.json -f web-Google.mtx -s=1 ```
Index
| Command | Action | | --------------- | ---------------------------- | | pagerank | Finds pagerank of all vertices. | | sssp | Finds single source shortest path from source vertex. | | traversal-bfs | Traverses breadth-first from source vertex. | | triangle-count | Counts triangles in undirected, lower triangular graph. |
References
- nvGraph pagerank example, EN605.617, JHU-EP-Intro2GPU
- nvGraph pagerank example, CUDA Toolkit Documentation
- nvGraph Library User's Guide
- RAPIDS nvGraph NVIDIA graph library
- Get path of current script when executed through a symlink
Owner
- Name: nodef
- Login: nodef
- Kind: organization
- Website: https://nodef.github.io/
- Repositories: 119
- Profile: https://github.com/nodef
A summary of programs made with Node.js.
Citation (CITATION.cff)
cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
- family-names: Sahu
given-names: Subhajit
orcid: https://orcid.org/0000-0001-5140-6578
title: "nodef/nvgraph.sh: CLI for nvGraph, which is a GPU-based graph analytics library written by NVIDIA, using CUDA"
version: 1.2.15
doi: 10.5281/zenodo.6647366
date-released: 2022-06-07
GitHub Events
Total
- Push event: 1
Last Year
- Push event: 1
Committers
Last synced: almost 3 years ago
All Time
- Total Commits: 52
- Total Committers: 1
- Avg Commits per committer: 52.0
- Development Distribution Score (DDS): 0.0
Top Committers
| Name | Commits | |
|---|---|---|
| Subhajit Sahu | w****7@g****m | 52 |
Issues and Pull Requests
Last synced: 4 months ago
All Time
- Total issues: 0
- Total pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Total issue authors: 0
- Total pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels
Packages
- Total packages: 1
-
Total downloads:
- npm 51 last-month
- Total dependent packages: 1
- Total dependent repositories: 0
- Total versions: 77
- Total maintainers: 1
npmjs.org: nvgraph.sh
CLI for nvGraph, which is a GPU-based graph analytics library written by NVIDIA, using CUDA.
- Homepage: https://github.com/nodef/nvgraph.sh#readme
- License: MIT
-
Latest release: 1.3.2
published 9 months ago
