Accelerating geospatial computing using Apache Arrow

on YouTube ↗04m

" summary

geoarrow extends Apache Arrow to store spatial types columnarly and runs spatial queries on the Arrow compute engine, demonstrating geoarrow::read_geoparquet() returning sf objects and parquet files that load much faster and keep types. It opens many files as an Arrow dataset (S3-hosted or local), processes queries batchwise and in parallel across cores so datasets bigger than memory complete, and passes results zero‑copy as Arrow arrays to Rust, Python, or R — package under development but working.

AI-written summary · about

" related talks

• 05mMake Big Geospatial Data Accessible with ArrowCari Gostic · posit::conf(2025)
• 01h 02mApache Arrow: High-Performance Columnar Data FrameworkWes McKinney · Vaccination Database Talks (Second Dose) 2021
• 33mArrow in Flight - New Developments in Data ConnectivityDavid Li · OSA Con 2022
• 45mApache Arrow: In Pursuit of SpeedIgor Seliverstov · JPoint 2022
• 46mEfficient Data formats for Analytics with Parquet and ArrowJulien Le Dem · Berlin Buzzwords 2016
• 32mVectorized Query Processing for CPUs using Apache ArrowJacques Nadeau · Scale By The Bay 2019
• 33mA 101 in time series analytics with Apache Arrow, Pandas and ParquetZoe Steinkamp · PyBay 2024
• 39mTaming the language border (...)Uwe L. Korn · Berlin Buzzwords 2019