This talk was recorded at NDC Sydney in Sydney, Australia. #ndcsydney  #ndcconferences #developer #softwaredeveloper    

Attend the next NDC conference near you: 
https://ndcconferences.com
https://ndcsydney.com/

Subscribe to our YouTube channel and learn every day:   
/   @NDC 

Follow our Social Media!

https://www.facebook.com/ndcconferences
https://twitter.com/NDC_Conferences
https://www.instagram.com/ndc_conferences/

#database #bigdata #devops 

Scaling a database sounds like a great idea, but have you considered the number of ways it can fail? Add some nodes, throw in replication, and suddenly you're explaining to stakeholders why two customers saw two different truths at the same time. Turns out distributed systems have opinions about physics - but data modeling can help.

This session covers the fundamentals (replication, consistency, partitions) and the tradeoffs nobody mentioned when they said "just scale horizontally." We'll look at real failure modes and practical patterns for building systems that handle chaos without dragging your app down with them.

NDC Conferences

A conference talk covering how distributed databases fail in practice and the trade-offs developers must understand. Starting with a simple notebook analogy, it explains replication lag, CAP theorem, network partitions, and sharding pitfalls. Real-world incidents (DynamoDB DNS outage, Cloudflare permissions bug) illustrate blast radius risks. Five practical patterns are presented: choosing good shard keys, idempotency keys, the transactional outbox pattern, sagas for multi-service transactions, and denormalization for read performance. Closes with five diagnostic questions developers should be able to answer before running distributed systems in production.

Distributed Databases: What Could Possibly Go Wrong? - Heather Downing - NDC Sydney 2026