Blog
FlatQL blog
Notes on how the engine works and why it is built this way.
Exporting 5 million rows from Elasticsearch without OOM
Why from/size and buffered results fail on large Elasticsearch exports, how FlatQL streams scroll pages to a file on disk, and why the cap is 5,000,000 rows.
Per-index access control for Elasticsearch without creating cluster users
Give each engineer read or write access to specific Elasticsearch indices through one shared connection, checked before any request reaches the cluster.
Pushdown explained: what runs on the cluster vs the engine
How FlatQL decides what Elasticsearch runs and what the embedded engine runs: the exact rules, three worked plans, and how to rewrite statements so more pushes.
Why Elasticsearch SQL can't do joins (and how FlatQL does)
Elasticsearch SQL reads one index per statement and has no server-side join. Here is why, what the alternatives cost, and how FlatQL joins indices anyway.
Start on Free. It stays free.
Connect a cluster and run a join in five minutes. No card, no expiry.