You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes
Tech · Releases · Mon, September 21, 2026
You can now use the AlloyDB Columnar Engine as a read-optimized, in-memory cache for HNSW vector indexes. This feature is generally available ( GA ). It accelerates vector search performance and increases queries per second (QPS) for vector
Every year
Also on Sep 21, 2026
| Date | Event | When |
|---|---|---|
| Mon, Sep 21, 2026 | Yom Kippur | 4 days ago |
| Mon, Sep 21, 2026 | 250th anniversary: American Revolutionary War: The Great Fire of New York (depicted) broke out... | 4 days ago |
| Mon, Sep 21, 2026 | International Day of Peace | 4 days ago |
| Mon, Sep 21, 2026 | Gartner Data & Analytics Summit Mumbai 2026 | 4 days ago |
Source, method and history
- Source
- google-cloud
- How we know
- From the Releases API (releases.fru.dev), which reads each vendor's changelog.
- Who
- google-cloud
- Recurs
- one-off
- More on fru.dev
- releases.fru.dev
- Feed
- Subscribe to Releases
- Changes
- 2026-09-25: added to 2026-09-21