Databricks Geospatial Analytics at Scale Wherobots is a spatial analytics engine that runs on your Databricks data through Unity Catalog, with no data movement and no governance changes. Spatial queries run up to 20x faster at up to 60% lower cost than other managed Spark engines. 300+ spatial functions covering vector and raster data, with native Spark SQL for tabular operations. Start for Free Request a Demo Your Unity Catalog Governance Stays Intact Wherobots authenticates to Unity Catalog with a Databricks service principal (OAuth or PAT). Existing access controls, table permissions, and lineage carry over. No parallel governance, no data copy. See how Wherobots connects to Unity Catalog Read Delta and Iceberg Tables, Write Results Back, No Migration Required Wherobots reads Delta and Iceberg tables directly under Unity Catalog and writes results back to Unity Catalog or Wherobots Data Hub. No ETL into a separate spatial environment. Read the Delta and Iceberg integration docs Up to 20x better performance, unlimited scale, at 1/n the cost WherobotsDB is a Spark and Apache Sedona compatible engine with query planner optimizations built for spatial workloads. Overture Maps processed 2.6 billion building geometries up to 20x faster after redirecting Sedona jobs to WherobotsDB. GeoPostcodes cut 39 days of processing to under one day. Read the Overture Maps performance benchmark One Engine for Vector, Raster, and Tabular Data Most spatial tools handle vector or raster, not both. Wherobots runs 300+ spatial functions across vector and raster data, with native Spark SQL for tabular, in one engine. Join parcel boundaries against satellite imagery against a customer CSV without moving data between tools. See the GeoPostcodes case study Run Computer Vision Models on Satellite and Drone Imagery with RasterFlow RasterFlow, the Wherobots inference engine, runs computer vision models on satellite and drone imagery at any scale. Customers have detected airplanes in satellite imagery for $2.50 and processed millions of fleet-derived traffic signs for automated map updates. No separate ML stack required. Learn about SAM3 and Rasterflow Build in SQL, Python, or Scala, No New Toolchain Required Spatial SQL, Python, and Scala through the same notebook and API surface your team already uses. Existing Apache Sedona jobs run on WherobotsDB without code changes. New spatial capability, no new toolchain. Browse the Spatial SQL API reference “Overture’s 2.6B-building dataset now runs up to 20x faster on Wherobots, with a simple code redirect that kept Sedona compatibility and made our team more productive.” Jennings Anderson Software Engineer with Meta for Overture “With Wherobots on AWS, we can now scale to millions of acres reliably and cost-effectively—delivering faster results and more value to our customers.” G. Bailey Stockdale CEO Leaf Agriculture “With Apache Sedona, we process millions of fleet-derived traffic signs, using scalable spatial joins and partitioning to automate map updates—enhancing Amazon Last Mile’s delivery networks for faster, more reliable routing.” Arka Pratim Das Sr. Manager, Software Development, Amazon Maps “Getting data, algorithms, and compute in one place with Spark/Sedona notebooks is a huge boost—powerful like Earth Engine, but with the control developers need to get jobs done.” John Powell Sr. Geospatial Data Engineer, AddressCloud “39 days of processing reduced to <1 day—delivering faster, more accurate population insights to our logistics and supply chain customers worldwide.” Jerome Urbain Head of Products at GeoPostcodes Built On Apache Sedona, Trusted by Fortune 500 Companies Use cases Spatial Analytics Use CasesBuilt for Scale Spatial Joins on Millions of Records for Under 50 Cents Perform spatial joins on millions of records for less than 50¢ RUN THIS NOTEBOOK GPS Map Matching at Scale Match GPS points to roads for millions of records with no APIs RUN THIS NOTEBOOK Detect Airplanes from Satellite Imagery with AI Use text prompts and AI to find airplanes in satellie imagery for $2.50 RUN THIS NOTEBOOK Frequently Asked Questions Does Wherobots work with Databricks Unity Catalog? Yes. Wherobots connects directly to Databricks Unity Catalog so you can read from and write to your existing Iceberg or Delta tables without moving data or creating a parallel governance environment. Your existing access controls and data governance policies stay intact. Do I need to migrate my data to use Wherobots with Databricks? No. Wherobots uses data federation to query your data in place inside Unity Catalog. There is no data copy and no migration required before you can run spatial queries. Can I run my existing Apache Sedona jobs on Wherobots without rewriting them? Yes. WherobotsDB is fully compatible with Apache Sedona and Spark. You can redirect existing jobs to run on WherobotsDB with zero code changes and immediately gain performance and cost improvements. How much faster is Wherobots than other managed Spark engines? Wherobots customers self-report 5x to 20x faster spatial query performance compared to other managed Spark engines and leading data platforms. Overture Maps processed 2.6 billion building geometries up to 20x faster after redirecting existing Apache Sedona jobs to WherobotsDB with no code changes. What spatial data formats does Wherobots support? Wherobots supports raster formats including GeoTIFF, Zarr, and NetCDF for satellite and sensor data, and vector formats including GeoParquet, Shapefiles, and GeoJSON for spatial geometries. Both data types are processable alongside standard tabular data in a single engine. Is Wherobots available on AWS? Yes. Wherobots is available on AWS and on the AWS Marketplace. Enterprise customers can also run Wherobots in their own AWS VPC for maximum data control and security compliance. Start Running Spatial Analytics on Your Databricks Data Spin up a free Wherobots workspace, connect Unity Catalog in minutes, and run your first spatial query on Delta or Iceberg tables. TRY NOW
