技巧精选
Databricks 实测 Lakebase Search：缓存大小比数据行数更影响向量搜索延迟
精选理由
Databricks 拿 1M 到 4M 条数据实测了向量搜索，发现缓存配比对延迟影响最大，做 RAG 的人可以直接抄结论。
Databricks 在 100 万到 400 万条 passage 上对 Lakebase Search 做了向量搜索基准测试。结果显示缓存大小与向量索引的比例比行数更关键：当缓存与索引规模大致相当时，95–98% 的页面读取可以避开存储层，热查询 p50 延迟保持在 11 到 23 ms 之间。数据集增大时，把缓存和向量索引同步扩容有助于维持低延迟。
原文 · Databricks
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://t.co/Bb06flLWjV
