Catalog
concept#Architecture#Integration#Platform#Reliability

Coexistence Architecture

An architectural approach for running legacy systems alongside new components and enabling incremental modernization.

Coexistence Architecture defines strategies for running legacy systems alongside new cloud‑native components.
Established
Medium

Classification

  • Medium
  • Technical
  • Architectural
  • Intermediate

Technical context

API gateway / service meshEvent bus / message brokerData replication tools (CDC)

Principles & goals

Decouple via clearly defined interfacesPrefer phased migration over big‑bang replacementMake operational boundaries and observability explicit
Build
Enterprise, Domain

Use cases & scenarios

Compromises

  • Inconsistent data states across systems
  • Hidden dependencies complicate extraction
  • Operational costs increase due to parallel operation
  • Use automated testing and canary releases
  • Establish clear integration contracts and versioning
  • Ensure observability across transitional boundaries

I/O & resources

  • Inventory of interfaces and dependencies
  • Operational requirements and SLAs
  • Data model and replication strategy
  • Defined integration layer and transitional architecture
  • Reduced monolith functionality
  • Metrics for consistency, availability and cost

Description

Coexistence Architecture defines strategies for running legacy systems alongside new cloud‑native components. It emphasizes interface layers, decoupling, data consistency and phased migration so operations and development can coexist. The goal is to reduce risk, enable continuous delivery and achieve incremental modernization without full replacement.

✔

  • Reduces migration risk through incremental changes
  • Enables parallel operation and continuous delivery
  • Preserves investments in existing systems

✖

  • Requires additional integration and operational effort
  • May introduce short‑term complexity and redundancy
  • Not all components are suitable for phased separation

  • Mean Time to Migrate (MTTM)

    Average time to move a feature from the legacy system into the new architecture.

  • Data inconsistency incidents

    Count and severity of incidents involving inconsistent data states.

  • Operational cost of parallel systems

    Monthly cost for running legacy and new components in parallel.

Strangler pattern in legacy webshop

An online shop incrementally extracts checkout functionality into new services while the old shop remains running.

Hybrid cloud for core ERP

Critical ERP modules remain on‑premise while analytics run in the cloud with synchronized interfaces.

Event bridge for data coupling

Event‑based replication provides near‑real‑time consistency between the legacy system and new microservices.

1

Analyze: create interface inventory and dependency map

2

Design: define integration layer and migration paths

3

Pilot: extract a small domain and implement monitoring

4

Iterate: capture lessons learned and expand incrementally

⚠️ Technical debt & bottlenecks

  • Temporary adapter layers are not removed
  • Short‑term data copies persist permanently
  • Unmanaged interface versions lead to fragmentation
Data consistencyNetwork latencyLegacy dependencies
  • Parallel operation is prolonged indefinitely and increases costs
  • Data is incrementally duplicated without conflict resolution
  • Legacy system remains unchanged while new components act only as a façade
  • Underestimating hidden dependencies
  • Missing automation for tests and deployment
  • Unclear responsibilities between teams
Architecture knowledge on integration patternsOperations and cloud know‑howExperience with data migration and consistency strategies
Minimize downtime during migrationCompliance with regulatory data requirementsEnsure interoperability between components
  • • Existing proprietary interfaces may limit migration
  • • Operational SLAs must be maintained during coexistence
  • • Compliance and data protection requirements constrain data flows