From Monolithic Databases to Distributed SQL - Running Active-Active Across Regions!

If you're running traditional SQL workloads, you know the challenges: scaling beyond a single server means complex replication setups, sharding nightmares, or expensive hardware upgrades. What if you could get global-scale distributed SQL with true active-active multi-region capabilities without abandoning the relational model? This session introduces a new approach to distributed SQL databases - Amazon Aurora DSQL that maintains ACID guarantees while scaling horizontally across multiple regions. You'll learn how to build modern serverless applications that automatically scale to handle unpredictable workloads without capacity planning, using familiar SQL syntax and standard database drivers.

We'll explore practical implementation patterns for creating applications that scale from zero to millions of requests, examining real-world performance characteristics and the cost model of consumption-based pricing. The session covers critical decision factors: when to choose distributed SQL over traditional clustered databases. We'll discuss the trade-offs - including retry logic, optimistic concurrency patterns, and transaction scope considerations that differ from traditional single-node databases. Whether you're modernizing legacy applications or building new cloud-native systems, you'll gain practical insights for choosing the right database architecture and understanding what it really takes to implement it successfully.

Understand how Amazon Aurora DSQL enables globally distributed, active-active SQL with ACID guarantees.

Learn practical patterns for building serverless applications that scale automatically without capacity planning.

Evaluate when distributed SQL is the right choice and the trade-offs vs traditional databases.