Reliability
The reliability pillar encompasses the ability of a workload to perform its intended function correctly and consistently when it’s expected to.
Each best practice in this section is presented based on its place in the ML lifecycle as detailed in Well-Architected machine learning lifecycle.
Lifecycle phases
