AWS adds foreign-key constraints to Aurora DSQL
The change addresses a known adoption objection for brownfield environments.
AWS has added foreign-key constraints to Aurora DSQL, including actions such as CASCADE and SET NULL, closing a gap that had discouraged some users from adopting the database. Rather than locking tables, the system checks relationships against a transaction snapshot and detects conflicting row changes at commit; a constraint violation ends the transaction with a serialization failure. AWS warns that writes involving linked tables require extra reads and recommends benchmarking workloads before adding constraints. The missing capability drew criticism when the service debuted at re:Invent 2024, while Luc van Donkersgoed said the addition makes DSQL more viable for brownfield systems. AWS has also added CloudWatch Database Insights for monitoring performance at the statement and cluster levels.
Why it matters
For organizations with existing systems, the feature removes a reason to rule DSQL out, making it more viable to consider for brownfield use. Operators also gain a way to investigate performance at both statement and cluster levels.
Signal or noise?
Does this story matter, or is it hype? Decide before you see what everyone else thinks.
Sources
- InfoQ