When designing distributed services, there are three properties enterprises commonly desire: 1) consistency 2) availability, and 3) partition tolerance. It is impossible to achieve all three. You might have already encountered the situation when asking for RTO (Recovery Time Objective) and RPO (Recovery Point Objective) and people answer they don't want any data loss or downtime. No one wants to lose data or access to their database service, but should you increase your architecture complexity, you might increase the risk of failure!

So, what is the right balance for you?

Join our Ask the Expert session where we will explore the CAP and PACELC theorems, what they mean for RDBMS and what tradeoffs need to be made to optimize availability and minimize data loss. Afterwards, we'll explore several real-world practical use cases that demonstrate these theorems and classify their data loss and availability.

Key takeaways:

• Maintaining high availability and minimizing data loss in PostgreSQL?
• Does “Zero” data loss exist?
• How do you figure out what your RPO is?
• How do you figure out what your RTO is?
• What is the ideal architecture for maintaining uptime and avoiding data loss?
• What are the CAP and PACELC theorems?

Speakers:
Host: Kevin Li, Director Sales Engineer EMEA, EDB
Speaker: Lætitia Avrot, Field CTO, EDB