Databricks benchmarks Lakebase Search vector latency at 1M to 4M passages

AIDatabricks benchmarked Lakebase Search on 1M to 4M passages and found cache size relative to the vector index mattered more than row count. When the cache roughly matched the index, 95–98% of page reads avoided storage and warm vector search p50 latency stayed between 11 and 23 ms.

DatabricksOfficial

@databricks

View on X

We benchmarked Lakebase Search on 1M to 4M passages to understand how vector search performs as the index grows. Here's what we learned 👇

Cache size relative to the vector index mattered more than row count. When the cache roughly matched the index, 95–98% of page reads avoided storage and warm vector search p50 latency stayed between 11 and 23 ms.

As datasets grow, sizing the cache alongside the vector index helps maintain low-latency search.
https://www.databricks.com/blog/sizing-lakebase-search-lessons-real-app-and-4-million-passages?utm_source=twitter&utm_medium=organic-social

Source details

Databricks · x.comPublished Nov 3, 3:38 PM ET · added here 3:48 PM ET

Original title: We benchmarked Lakebase Search on 1M to 4M passages to understand how vector search performs as the index grows. Here's what we learned 👇

Databricks benchmarks Lakebase Search vector latency at 1M to 4M passages

Topics

Tags