Terminology

2185 terms, tools, patterns and metrics an architect is expected to use precisely. Each one gets a short explanation of what it is, and — where it matters — what it is commonly confused with. Search filters as you type; the column headers sort.

All areas2185 Architecture Fundamentals77 Distributed Systems107 Data Architecture110 Cloud Architecture89 Networking88 API & Integration Architecture82 Reliability & Resilience75 Observability70 Performance & Capacity Engineering72 Security Architecture81 Cost Architecture & FinOps66 Business Architecture67 Architecture Communication67 Enterprise Architecture66 Legacy Modernization66 AI-Era Architecture69 Software Architecture & Engineering71 Architecture Patterns71 Architecture Decision-Making63 The Architect's Meta-Skills61 Delivery & Release Engineering66 Platform Engineering & Developer Experience70 Testing & Quality Architecture66 Data Platform Architecture64 Streaming & Real-Time Data69 Data Governance & Semantics69 Frontend & Experience Architecture66 Edge, Mobile & IoT68 Regulatory & Data Protection Architecture65 Assurance, Audit & Model Risk64

Kind Any concept785 practice729 pattern362 metric125 case-study77 tool62 protocol45

34 terms shown.

Term	Kind	Topic	What it is
Application Decomposition	practice	Application Decomposition	Breaking a large application into parts — where finding the seam matters more than the extraction technique.
Business-Cycle Observation Window Full-Cycle Shutdown Soak, Longest-Period Consumer Watch	practice	Decommissioning	Setting the period over which you watch for remaining consumers from the system's longest business cycle rather than from a round number of days, because a 30-day watch cannot see an annual job.
Characterisation Test Golden Master Test, Approval Test, Behaviour Snapshot	practice	Application Decomposition	A test that records what a system currently does - bugs included - rather than what it should do, providing the safety net that makes changing an untested legacy system possible at all.
Cutover Planning	practice	Cutover Planning	The plan for the moment of switching — including the decision points, the rollback, and the rehearsal that makes it survivable.
Cutover Runbook	practice	Cutover Planning	The step-by-step plan for a migration event, including verification points, decision criteria and the rollback path, rehearsed before it is used.
Data Migration Strategies	practice	Data Migration Strategies	Moving data between systems with different models — where the difficulty is meaning and quality rather than transfer.
Database Migration	practice	Database Migration	Moving data from one store to another while the system keeps running — where verification and reversibility matter more than the copy itself.
Database Migration Strategy	practice	Database Migration	The approach for moving data to a new store, which is usually the longest pole and the highest risk in any modernisation.
Decommissioning	practice	Decommissioning	Actually switching off and removing a replaced system, which is where the savings in every modernisation business case live.
Decommissioning Discipline	practice	Decommissioning	The work of actually switching a legacy system off, which is where migration savings are realised and which is routinely left undone.
Decomposition by Capability	practice	Application Decomposition	Breaking a monolithic system into parts along business capability lines rather than technical layers, so that each part changes for one reason.
Disposition Decision	practice	The Six Rs	Choosing one of the standard migration treatments for each application, on the basis of its value, condition and the constraint that is actually binding.
Explained-Difference Criterion Shadow Exit Criterion, Zero Unexplained Differences	practice	Parallel Run	Ending a shadow or parallel run when every observed difference has been explained, rather than when differences reach zero - because zero is unachievable and unexplained differences are by definition the dange…
Legacy Assessment	practice	Legacy Assessment	Establishing what a legacy system actually does, what depends on it, and what it costs, before deciding what to do with it.
Legacy Headroom Budget Legacy Capacity Allocation, Integration Throughput Quota	practice	Legacy Integration Patterns	The measured spare throughput of a legacy system after retry amplification, batch overlap and organic growth are deducted, allocated to new consumers as enforced quotas rather than published as available capacity.
Legacy System Assessment	practice	Legacy Assessment	A structured evaluation of a system's business value, technical condition and risk, used to decide what to do with it rather than to describe it.
Mainframe Modernisation	practice	Mainframe Modernization	Replacing or reducing dependence on a core system that is reliable, undocumented, and central to the business.
Migration Risk	practice	Migration Risk	The specific risks migrations carry, and the controls that reduce each — with reversibility as the master control.
Migration Risk Register	practice	Migration Risk	A maintained record of what could go wrong in a migration, with likelihood, impact, owner and mitigation, reviewed as the programme progresses.
Modernisation Business Case	practice	Modernisation Business Case	Justifying modernisation work, where the benefits are diffuse, the costs are concrete, and "it is old" persuades nobody.
Modernisation Value Case	practice	Modernisation Business Case	The argument that funds a modernisation programme, built on quantified business consequence rather than on technical obsolescence.
Move Group Migration Affinity Group, Move Wave Affinity Set	practice	Rehosting	The set of workloads that must cross to the new infrastructure in the same wave because the round trips between them sit on a user-facing critical path, derived from observed network flows rather than from the…
Ownership-First Decomposition Assign Owners Before Splitting, Logical Before Physical	practice	Database Migration	Assigning each table a single owning team before splitting any code or data, because the ownership step costs nothing, removes the worst of the coupling immediately, and makes every subsequent step tractable.
Parallel Run	practice	Parallel Run	Operating the old and new systems simultaneously on the same inputs and comparing outputs, to build evidence that the replacement is correct before relying on it.
Point of No Return Rollback Horizon, Commit Point	practice	Cutover Planning	The step in a migration after which rollback would lose data or state - identified in advance, with a named authoriser and a stated recovery plan beyond it.
Progressive Shutdown Graceful Decommissioning, Staged Retirement	practice	Decommissioning	Retiring a system through reversible stages - read-only, then blocked, then stopped, then removed - so each stage surfaces remaining consumers safely instead of breaking them.
Refactor and Re-architect	practice	Refactoring & Re-architecting	Changing the application's structure — the most expensive migration strategy, justified only where the structure itself is the constraint.
Rehost Lift and Shift	practice	Rehosting	Moving a workload to new infrastructure with no change to the application — fastest, cheapest to execute, and the source of the bill that discredits migrations.
Replatform	practice	Replatforming	Moving a workload with targeted changes that capture most of the platform's benefit without redesigning the application.
Retire-First Portfolio Bias Cheapest Sufficient Option, Retire and Retain Default	practice	The Six Rs	Assigning every application the cheapest modernisation option that satisfies the business driver, with a written justification required for anything more - which resolves most portfolios into retire and retain…
Six Rs Decision Migration Disposition	practice	The Six Rs	Classifying each application into rehost, replatform, refactor, repurchase, retire or retain, based on value and effort rather than on preference.
Traffic Shadowing Dark Launch, Mirroring, Shadow Traffic	practice	Strangler Fig	Sending real production traffic to a new implementation alongside the old one, serving the old response and comparing - the only reliable way to find undocumented behaviour before it matters.
Zero-Downtime Cutover	practice	Zero-Downtime Migration	Moving a live system to a new implementation or datastore without an outage, using dual writes, backfill and a reversible traffic switch.
Zero-Downtime Deployment in Practice Rolling Deployment, Online Schema Change	practice	Zero-Downtime Migration	Shipping without a maintenance window, and the version-skew constraint that governs every schema and contract change.