Catalog
concept#Architecture#Software Engineering#Governance

Architecture

Fundamental model for a system's structure, relationships and guiding principles to steer technical and business decisions.

Architecture describes the fundamental structure, components, and relationships of a system and the principles guiding their design.
Established
High

Classification

  • High
  • Technical
  • Architectural
  • Advanced

Technical context

CI/CD pipelines and deployment toolsMonitoring and observability platformsIdentity and access management systems

Principles & goals

Clear separation of responsibilities and explicit interface definitions.Decisions are context-dependent and should reflect quality attributes.Architecture is evolvable: iterative refinement and regular reviews.
Discovery
Enterprise, Domain, Team

Use cases & scenarios

Compromises

  • Overengineering leads to unnecessary complexity.
  • Lack of alignment with operational realities creates technical debt.
  • Unclear responsibilities prevent fast decision-making.
  • Document decisions concisely and traceably as ADRs.
  • Prioritize quality attributes and measure them with SLOs.
  • Conduct regular architecture reviews with cross-functional teams.

I/O & resources

  • Business goals and use cases
  • Technical constraints and existing systems
  • Team skills and operating model
  • Architecture decision records (ADR)
  • Component and interface diagrams
  • Quality goals with metrics and SLOs

Description

Architecture describes the fundamental structure, components, and relationships of a system and the principles guiding their design. It links business requirements with technical decisions, aligns quality attributes and interfaces, and provides a decision framework for scalability, security, and maintainability across organizational contexts.

✔

  • Improved scalability and predictability of system behaviour.
  • Clear decision basis for technology and organizational questions.
  • Easier maintainability and team coordination through responsibility boundaries.

✖

  • Can become overly bureaucratic if governance is missing.
  • Not every architectural recommendation fits every organizational context.
  • High initial effort for documentation and alignment.

  • Response time (P95)

    Measures the 95th percentile response time of relevant endpoints.

  • Availability (uptime)

    Percentage of time the system is reachable for users.

  • Mean Time to Recovery (MTTR)

    Average time to recover after incidents.

E-commerce platform with scalable services

Split into checkout, catalog and payment services with asynchronous communication and clear SLAs.

Banking system with hardened security architecture

Multi-layer security controls, separation of sensitive data and architecture-level auditing.

IoT platform with edge and cloud components

Sensitive processing at the edge, aggregation in the cloud and resilience via local caching strategies.

1

Stakeholder workshops to gather requirements

2

Create architecture decisions and diagrams

3

Iterative implementation with reviews and tests

⚠️ Technical debt & bottlenecks

  • Short-term workarounds instead of robust interface designs.
  • Unclear ownership leads to unresolved architectural issues.
  • Missing automation of tests and deployments
database IOnetwork latencyteam capacity
  • Introducing complex microservice architecture for trivial applications.
  • Strict standardization without considering local requirements.
  • Documentation that is outdated and does not match implementation.
  • Choosing a technology too early before clarifying requirements.
  • Neglecting non-functional requirements when making design decisions.
  • Insufficient involvement of operations and security in architecture decisions.
System architecture and domain modellingNetworking and security fundamentalsExperience with cloud or operations platforms
Scalability and performance requirementsSecurity and compliance requirementsTime-to-market and operational costs
  • • Budget and infrastructure limits
  • • Regulatory requirements for data residency
  • • Legacy systems with limited modifiability