Add regions to your database: `ALTER DATABASE mydb ADD REGION 'us-east1'; ADD REGION 'eu-west1';` — regions must match the cluster's node regions.
Set the primary region: `ALTER DATABASE mydb SET PRIMARY REGION 'us-east1';`.
For tables with a clear home region, use REGIONAL BY ROW: add a `crdb_region` column and `ALTER TABLE mytable SET LOCALITY REGIONAL BY ROW AS crdb_region;`.
For globally shared reference tables, use: `ALTER TABLE config SET LOCALITY GLOBAL;` — reads are served locally in all regions.
Verify placement with `SHOW CREATE TABLE mytable;` and use `EXPLAIN` on queries to confirm reads are served locally.
Known gotchas
REGIONAL BY ROW requires the partition key (`crdb_region`) to be included in the primary key; plan your schema accordingly before inserting large datasets.
GLOBAL tables replicate all data to all regions — only use for small, read-heavy reference tables to avoid amplified write costs.
Adding a region to a large database triggers a background rebalance that can temporarily increase cluster load.
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp