cugraph
Scalix
Our great sponsors
cugraph | Scalix | |
---|---|---|
6 | 1 | |
1,573 | 7 | |
2.8% | - | |
9.6 | 7.0 | |
3 days ago | 4 days ago | |
Cuda | Cuda | |
Apache License 2.0 | BSD 3-clause "New" or "Revised" License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
cugraph
- CuGraph – GPU-accelerated graph analytics
-
GPU implementation of shortest path?
cuGraph does some of what Networkx does, but it is far from being as easy to use. But it should be fast.
- NetworkX 3.0 has been released
-
GraphBLAS
https://en.wikipedia.org/wiki/Sparse_matrix :
> The concept of sparsity is useful in combinatorics and application areas such as network theory and numerical analysis, which typically have a low density of significant data or connections. Large sparse matrices often appear in scientific or engineering applications when solving partial differential equations.
CuGraph has a NetworkX-like API, though only so many of the networkx algorithms are CUDA-optimized.
https://github.com/rapidsai/cugraph :
> cuGraph operates, at the Python layer, on GPU DataFrames, thereby allowing for seamless passing of data between ETL tasks in cuDF and machine learning tasks in cuML. Data scientists familiar with Python will quickly pick up how cuGraph integrates with the Pandas-like API of cuDF. Likewise, users familiar with NetworkX will quickly recognize the NetworkX-like API provided in cuGraph, with the goal to allow existing code to be ported with minimal effort into RAPIDS.
> While the high-level cugraph python API provides an easy-to-use and familiar interface for data scientists that's consistent with other RAPIDS libraries in their workflow, some use cases require access to lower-level graph theory concepts. For these users, we provide an additional Python API called pylibcugraph, intended for applications that require a tighter integration with cuGraph at the Python layer with fewer dependencies. Users familiar with C/C++/CUDA and graph structures can access libcugraph and libcugraph_c for low level integration outside of python.
/? sparse
-
[D] Seeking Advice - For graph ML, Neo4j or nah?
I feel like you would need to develop a custom solution which might in part store data in Neo4j but you will have to figure out how to efficiently pull the data you need to train your GNNs; and I think this tends to be the bottleneck since Graph DBs are not optimised for the kinds of queries you need for GNNs. For what it's worth, I wouldn't really bother with implementing a custom graph data structure (unless I was really keen) as there are some good implementations out there. Have you looked at cuGraph for example?
- WSL2 CUDA/CUDF issue : Unable to establish a shared memory space between system and Vram
Scalix
-
Scalix: A Data Parallel Compute Framework w/ Automatic Scaling
The repo can be found here: https://github.com/NAGAGroup/Scalix
What are some alternatives?
pygraphistry - PyGraphistry is a Python library to quickly load, shape, embed, and explore big graphs with the GPU-accelerated Graphistry visual graph analyzer
TorchPQ - Approximate nearest neighbor search with product quantization on GPU in pytorch and cuda
Memgraph - Open-source graph database, tuned for dynamic analytics environments. Easy to adopt, scale and own.
monolish - monolish: MONOlithic LInear equation Solvers for Highly-parallel architecture
rmm - RAPIDS Memory Manager
cub - [ARCHIVED] Cooperative primitives for CUDA C++. See https://github.com/NVIDIA/cccl
mage - MAGE - Memgraph Advanced Graph Extensions :crystal_ball:
MegBA - MegBA: A GPU-Based Distributed Library for Large-Scale Bundle Adjustment
demo-news-recommendation - Exploring News Recommendation With Neo4j GDS
ArrayFire - ArrayFire: a general purpose GPU library.
graph-data-science - Source code for the Neo4j Graph Data Science library of graph algorithms.
instant-ngp - Instant neural graphics primitives: lightning fast NeRF and more