Reliability pillar
The reliability pillar encompasses the ability of a service to perform its intended function correctly and consistently when it’s expected to. This includes the ability to operate and test the service through its total lifecycle and includes a high level of planning around service continuity and perceived reliability by the public.
The following questions and best practices are designed to complement the best practices in the Reliability Pillar whitepaper.
| GL-REL-01: How do you test the service and demonstrate resilience? |
|---|
-
Establish and document test-driven design and implementation, including scenario testing and mitigation planning: Establish system and process test environments, including scaled functional testing, user testing with the public, war gaming for procedural and scenario testing, and the potential for regulatory sandboxes for policy testing.
-
Improvement plan – Consider running regular AWS Well-Architected game days in the lead up to launching a new product to identify areas for improvement and end to end system/service reliability. Encourage the organization to consider persistent team structures that span policy and delivery in their forward planning for the service, design and run scenarios and mitigation planning sessions with the customer and scaled test processes and planning.
-
-
Seek peer review of reliability planning: Run and document feedback and recommendations from peer review sessions for the service and reliability planning.
-
Improvement plan – Support the customer to engage with relevant peers to review, share, and help assure the best possible reliability of the service.
-
| GL-REL-02: How do you plan for service continuity? |
|---|
Continuity of service is one of the highest priorities for government services. Many government services are not optional for people. These services are relied upon when we are at our most vulnerable. Business continuity is especially important to verify that end users are not put at risk from a lack of access to critical government services and systems.
-
Assess the impact of downtime and service deadlines: Test and document the assumptions about the likely impacts of service downtime, especially on vulnerable people and communities, and around anticipated deadlines such as elections, seasonal emergencies (such as fire season), and waiting times at medical care facilities. Capture these assumptions in user journey maps and document remediations and strategies in the concept of operations document.
-
Improvement plan – Analyze and document the full impacts of downtime on end users, stakeholders, and others who depend on the service, using a variety of relevant planned and unplanned scenarios.
-
-
Plan for a graceful degradation of service, with appropriate fail-overs and alternative service pathways: Verify that the service has appropriate fail-over strategies proportionate to the risk, and that users can identify alternative pathways to the service, including offline and phone-based user journeys. Verify that alternative service pathways are quickly discoverable by end users, even if the digital services are down. Have a stakeholder list, FAQs, and a communications plan prepared.
-
Improvement plan – Use scenario planning, user journey maps, and support the organization to resource and plan for service degradation, failovers, and alternative pathways for the service.
-
-
Verify and document the processes and public reporting of post-incident reviews: Government departments often have specific requirements to report incidents publicly, and even when there isn't a requirement, it is often good practice to help maintain high public trust and confidence in the public institution. Identifying any requirements and ensuring the processes and expectations are clear and documented can assist in future incident management.
-
Improvement plan – Work with the organization to document any specific or legislative requirements to report incidents publicly, and document relevant processes.
-
Resources
