Speaker: Jason Barto

Will my system keep working when a server fails? When a data center goes offline? When a service dependency is unavailable?

Availability calculations for redundant components require that those components are independent and autonomous of each other. But modern day systems are complex, exhibiting unexpected behaviours, and what was thought to be autonomous may in fact be indirectly dependent. 

Fault isolation boundaries give us a way to think about system design and understand relationships between system components. Chaos engineering gives us a way to test this autonomy and validate that our systems are implemented as designed, building confidence in the system’s capability to withstand turbulent conditions.

In this session we will talk about fault isolation boundaries and ways to take advantage of fault isolation in AWS. We will then demonstrate initial tests you can use to ensure your system has successfully isolated faults within its architecture.

This panel will aim to explore, share ideas and provide pragmatic insight around some key areas related to designing, running and maintaining resilient architectures.