Are Stored Procedures and Triggers Anti-Patterns in the Cloud Native World?

The rise of cloud-native applications has led to a shift in how business logic is implemented. While some newer distributed SQL databases like Google Spanner and CockroachDB have chosen not to implement traditional features such as triggers, stored procedures, and expression indexes, others like Amazon Aurora and YugabyteDB continue to support these features by reusing existing RDBMS engines. The importance of these features for cloud-native applications was highlighted through practical examples, including automatic timestamping with triggers, efficient geo-distributed batch updates using stored procedures, data validation in GraphQL/JAMStack applications using user-defined functions, case-insensitive lookups with expression indexes, server-side cursors for scanning large tables, row-level security (RLS) for multi-tenant data isolation, and async replication for modernizing apps. These features can simplify application development while improving performance and strengthening security, making them essential for cloud-native applications built atop distributed SQL databases.