GeoPython 2024: Fast geospatial data visualization and analysis with Lonboard and GeoArrow

Kyle Barron
Fast geospatial data visualization and analysis with Lonboard and GeoArrow
Lonboard and geoarrow-rust are two new libraries that leverage the GeoArrow data format for speeding up geospatial data visualization and analysis. Learn how the GeoArrow ecosystem can speed up your geospatial Python workloads.
Lonboard is a new Python library for fast, interactive geospatial vector data visualization in Jupyter. By building on cutting-edge technologies like GeoArrow and GeoParquet in conjunction with GPU-based map rendering, lonboard can visualize millions of coordinates up to 50x faster than existing libraries.
geoarrow-rust is a new fundamental Python library for efficient geospatial operations that connects to algorithms implemented in Rust - independent from GEOS and GDAL. It's fast! Early benchmarking has shown performance 2-5x faster than pyogrio/OGR for reading and writing selected geospatial file formats, and it has potential to significantly speed up operations like spatial joins.
This talk will discuss why these libraries are so fast, how the growing GeoArrow ecosystem interoperates so efficiently, and how you can use these tools in your own work.