Return to Well-Architected Framework Guide

Appendix: Reliability

How do you back up data?

• Identify and back up all data that needs to be backed up, or reproduce the data from sources
• Secure and encrypt backups
• Perform data backup automatically
• Perform periodic recovery of the data to verify backup integrity and processes

How do you use fault isolation to protect your workload?

• Deploy the workload to multiple locations
• Automate recovery for components constrained to a single location
• Use bulkhead architectures to limit scope of impact

How do you design your workload to withstand component failures?

• Monitor all components of the workload to detect failures
• Fail over to healthy resources
• Automate healing on all layers:
• Use static stability to prevent bimodal behavior
• Send notifications when events impact availability

How do you test reliability?

• Use playbooks to investigate failures
• Perform post-incident analysis
• Test functional requirements
• Test scaling and performance requirements
• Test resiliency using chaos engineering
• Conduct game days regularly

How do you plan for disaster recovery (DR)?

• Define recovery objectives for downtime and data loss
• Use defined recovery strategies to meet the recovery objectives
• Test disaster recovery implementation to validate the implementation
• Manage configuration drift at the DR site or region
• Automate recovery

Return to Well-Architected Framework Guide