AivexaNewsSearch
AI news for builders and product teamsChecked every hour

Geospatial queries are up to 38X faster with Lakehouse//RT

Collected Oct 15, 2026

Databricks announced that spatial SQL and spatial joins are now supported on Lakehouse//RT, and says those geospatial queries run up to 38X faster.

That is the extent of the published detail: the capability is spatial SQL plus spatial joins, on a runtime the company calls Lakehouse//RT, and the headline performance claim is a maximum speedup of 38X. Databricks has not, in this announcement, listed supported spatial functions, geometry or geography types, indexing behavior, supported regions or runtimes, or any date or pricing information.

For developers building mapping, logistics, mobility or location-analytics features, the practical question is whether spatial predicates can sit next to the rest of their data rather than being pushed to a separate geospatial store. This suggests Databricks is positioning Lakehouse//RT as a place where spatial joins execute close to the lakehouse data itself. Treat the 38X figure as a best-case number; speedups of that size usually reflect comparing a tuned native implementation against a general-purpose fallback, so your mileage will depend on data shape, join selectivity and workload.

Why it matters: teams that maintain a separate spatial database alongside their lakehouse may be able to fold geospatial workloads into their existing pipelines, which would reduce data movement and duplicated infrastructure. Whether that holds depends on details Databricks has not yet published, so I would treat this as a signal to benchmark rather than a reason to migrate today.

Read at Databricks

Based on reporting from the original publisher. Visit the source for full context and later updates.

Publisher excerpt

Spatial SQL and spatial joins now run on Lakehouse//RT and they're up to 38X faster...