If you're a database architect, you've already encountered the situation when

you ask for RTO and RPO and people answer they don't want any data loss and

downtime is not an option either.

If you discuss with people aware of the CAP theorem, they will temper their

demands by asking for 99.999% availability, emphasising they didn't ask for the

impossible 100%...

In this conf talk, we'll explore the CAP theorem, what it means for RDBMS and

what tradeoffs need to be made. Then we'll explore several practical use cases

and we'll try to classify their data loss and availability.

About the speaker:

Lætitia is a PostgreSQL expert in France.

She begun working with PostgreSQL in 2007. She also had to learn Oracle and SQL Server at the same time.

She had a lot of exciting experiences with high availability, production crashes, disaster recovery plan, load balancing, spatial data...