What is Operations Architecture (OA)
Operations Architecture or OA is often defined as the unified collection of technology services, tools, standards and controls required to monitor and manage business application environments to keep them operating at the defined service levels.
When I think of Ops Arch I use my go to list of themes:
- Production Scheduling
- Infrastructure Monitoring and Alerting
- Application Availability Monitoring & Alerting
- Performance Monitoring and Alerting
- Network Monitoring and Alerting
- Event Management (correlation and processing)
- SLA/OLA attainment
- Operational Controls (Interface Accounting)
- Operational Backup & Recovery
- Disaster Recovery
- Integration into the ITIL / SIAM environment
For every new project that delivers an IT Service or capability I will evaluate it against each heading. I want to check that the project not only delivers the business functionality and performance requirements but also delivers a sustainable Ops Arch solution for BAU monitoring and management.
Application monitoring can be further split into 3 sub-categories:
- Synthetic Monitoring
- Real User Monitoring
- Application Performance Monitoring
Synthetic monitoring is useful for baselining availability and performance without the unknown variables introduced by real user intervention. However you need to take care not to leave a footprint if data is extracted later for MI or other purposes.
Real User monitoring using a network port mirror or injecting code to capture the user experience is more reflective of how the users interact with your IT Services and can capture issues with usability as well as technical issues.
Application Performance Monitoring (APM) is the advanced monitoring technique for complex distributed applications. As more services are delivered using digital workspaces, containers and cloud, APM tracks the user transaction across platforms and can localise issues making incident triage a lot easier.
Good Operations Architecture should be reused across deliveries and be set up to meet the strategic goals of the Service & Operations department. Allowing each project to define the Ops Arch deliverables can lead to a piecemeal approach of disparate monitoring and alerting systems and islands of events.
Apply the same good rigour of architecture principles to your operational architecture.