pwrliang opened a new pull request, #774:
URL: https://github.com/apache/sedona-db/pull/774
## What this PR does
This PR adds a new documentation page at `docs/gpu-acceleration.ipynb` and
`docs/gpu-acceleration.md` that introduces GPU-accelerated spatial joins in
SedonaDB.
## Highlights
- Adds prerequisites for GPU usage (CUDA/version, compute capability, RT
core notes).
- Documents GPU join execution model (filtering + refinement) and supported
predicates.
- Explains fallback behavior with `gpu.fallback_to_cpu`.
- Provides build-from-source guidance for GPU
(`MATURIN_PEP517_ARGS="--features gpu"`), including related environment
variables.
- Adds runtime configuration examples for `gpu.*` options and batch-size
tuning.
- Includes a complete example workflow:
- check GPU availability (`nvidia-smi`)
- download sample data
- create external tables
- run CPU vs GPU benchmark in notebook-style code
- display timing/speedup output
## Why
Users need a practical, end-to-end guide to enable, tune, and validate GPU
spatial join performance.
## Scope
- Documentation only
- New file: `docs/gpu-acceleration.md`
- No runtime/engine behavior changes
## Notes
Benchmark numbers shown in the example are environment-dependent and should
be treated as illustrative.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
