Terminology
2205 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 areas2205
Architecture Fundamentals77
Distributed Systems107
Data Architecture110
Cloud Architecture94
Networking88
API & Integration Architecture82
Reliability & Resilience75
Observability70
Performance & Capacity Engineering72
Security Architecture86
Cost Architecture & FinOps71
Business Architecture67
Architecture Communication67
Enterprise Architecture66
Legacy Modernization71
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
368 terms shown.
| Term | Kind | Topic | What it is |
|---|---|---|---|
| Output Validation Layer | pattern | Guardrails | A deterministic check applied to model output before it is used, treating the model as an untrusted component. |
| Partial Hydration Islands Architecture, Selective Hydration | pattern | Hydration Cost | Attaching client-side interactivity only to the components that need it, rather than reconstructing the whole page's component tree in the browser. |
| Paved Roads Golden Path | pattern | Paved Roads | A supported, opinionated default path that is the easiest way to do the right thing — governance by convenience rather than by mandate. |
| Per-Dependency Concurrency Limit Dependency Semaphore, Downstream Bulkhead | pattern | Bulkheads & Isolation | A hard cap on in-flight calls to a specific downstream, which bounds the damage a slow dependency can do regardless of how slow it becomes - more reliable than a circuit breaker because it needs no detection. |
| Per-Entity Migration State Record-Level Ownership Flag, Entity Migration Flag, Per-Row Routing State | pattern | Coexistence Patterns | A field on each business entity naming the system currently authoritative for it, which makes the unit of migration one row rather than one capability - so traffic moves in nameable cohorts and rolls back by u… |
| Per-Record Tenancy Record-Level Ownership, Attribute-Based Tenancy, Cross-Org Data Ownership | pattern | Authorization | Attaching the owning organisation to each record rather than to its container, so that data shared across tenant boundaries can still be governed, exported, retained and deleted according to the policy of whoe… |
| Persisted Query | pattern | GraphQL | Registering allowed queries in advance and having clients send an identifier instead of the query text. |
| Phishing-Resistant Authentication Origin-Bound Authentication, Unphishable MFA | pattern | Authentication | Authentication whose response cannot be replayed at a site other than the one it was produced for, because the authenticator signs the requesting origin rather than releasing a secret the user could pass on. |
| Pilot Light | pattern | Disaster Recovery | A disaster recovery posture where core data is continuously replicated and minimal infrastructure runs, with the rest provisioned only on failover. |
| Pipeline Fan-Out | pattern | Pipeline Architecture | Splitting independent verification work to run in parallel, bounded by what genuinely has no ordering dependency on anything else. |
| Pipes and Filters | pattern | Pipes and Filters | Decomposing processing into independent stages connected by data channels, so each stage can be developed, tested, scaled and restarted alone. |
| Pipes and Filters | pattern | Architecture Patterns | Decomposing processing into independent steps connected by channels, each transforming its input and passing it on. |
| Policy Decision Point PDP, Authorization Service, Policy Engine | pattern | Zero Trust | The component that evaluates an authorisation question and returns a decision, kept separate from the enforcement points that ask, so policy can change without redeploying every service. |
| Port and Adapter | pattern | Hexagonal Architecture | An interface defined by the application expressing what it needs, paired with an implementation that connects it to a specific technology. |
| Pre-Provisioned Capacity | pattern | Static Stability | Running capacity that is already in place to absorb a failure, rather than depending on a control plane to create it during the failure. |
| Predicate-Based Filtering | pattern | Row & Column-Level Security | A policy that appends a filter to every query on a table based on who is asking, so one physical table serves many audiences safely. |
| Primary-Key-Partitioned Upsert Upsert Partitioning Contract, Key-Routed Upsert | pattern | Real-Time Analytical Stores | Routing every version of a record to one partition so a real-time store can resolve the latest value locally - which buys sub-second latest-value queries and fixes the partition count and key budget for the ta… |
| Priority Load Shedding Tiered Shedding, Criticality-Based Shedding, Selective Degradation | pattern | Load Shedding | Dropping traffic by declared business criticality under stress rather than uniformly or randomly, exploiting the fact that the highest-volume requests are usually the least valuable ones. |
| Priority Queueing | pattern | Load Shedding | Classifying requests by business importance so that overload sheds the least valuable work first rather than an arbitrary slice. |
| Process Manager | pattern | Orchestration vs Choreography | A component that owns the state and sequencing of a multi-step business process, deciding what happens next in response to events. |
| Progressive Media Materialisation Lazy Transcoding, On-Demand Rendition | pattern | Content Delivery Networks | Generating only a baseline media rendition on upload and producing higher-quality variants on first demand - trading a slower first view of unpopular content for a large reduction in transcoding and storage cost. |
| Projection | pattern | CQRS | The process that consumes changes from the write side and maintains a read model, and the component where most CQRS bugs live. |
| Publish-Subscribe Pub/Sub | pattern | Publish/Subscribe | One publication delivered to many subscribers, with the publisher unaware of who they are — and the fan-out problem that comes with it. |
| Publish/Subscribe Pub/Sub | pattern | Architecture Patterns | A messaging pattern where each published message is delivered to every interested subscriber, rather than to one competing worker. |
| Purpose-Based Access | pattern | Data Access Models | Granting access for a stated and recorded purpose rather than by role alone, which is what several data protection regimes actually require. |
| Query Federation | pattern | Data Virtualisation | Executing a single query across several underlying systems without moving the data, and the performance ceiling that pushdown determines. |
| Query-Based CDC Polling CDC, Timestamp-Based CDC | pattern | Change Data Capture | Detecting changes by repeatedly querying for rows modified since the last run — simple, universally available, and lossy in specific ways. |
| Rate Limiting | pattern | API & Integration | Bounding how many requests a caller may make in a window, to protect capacity and enforce fair use. |
| Rate Limiting in Practice Throttling, Quota Enforcement | pattern | Rate Limiting | Bounding how much work one caller can demand, the algorithm choice that shows through to users, and why distributed enforcement is approximate. |
| Read Local Write Global | pattern | Edge Data Consistency | Replicating read-only state to every edge location while directing writes to a single authoritative region, which is the arrangement most edge data stores actually provide. |
| Read Model Query Model, Projection Store | pattern | CQRS | A data structure shaped for a specific query rather than for the domain, maintained separately from the write model. |
| Read Model Projection | pattern | CQRS | A denormalised store built and maintained from write-side events specifically to serve one query shape efficiently. |
| Read Replica | pattern | Data Architecture | A copy of a database that receives changes from the primary and serves read-only queries, spreading read load. |
| Read-First Extraction Reads Before Writes, Reversible Extraction Order | pattern | Strangler Fig | Extracting a service's read path before its write path, because a wrong read migration is a flag flip while a wrong write migration is a data reconciliation exercise. |
| Real-Time Serving Layer | pattern | Real-Time Serving Layer | The read path that answers queries in milliseconds from precomputed or continuously updated state — where the design question is what to compute when. |
| Reconciliation Loop | pattern | GitOps | A controller that continuously compares declared state with actual state and acts to close the gap, replacing deployment as an event with convergence as a property. |
| Release Ring | pattern | Progressive Delivery | An ordered cohort of users a change reaches before the next one, chosen so that early rings would notice a fault and tolerate it. |
| Release Strategies | pattern | Release Strategies | How new code reaches users — rolling, blue-green, canary, or flag-controlled — chosen by rollback speed and blast radius. |
| Release Toggle Temporary Feature Flag, Rollout Flag, Deploy-Release Decoupler | pattern | Feature Flags | A short-lived flag that decouples deploying code from releasing a feature, intended to be removed within weeks - and the category responsible for almost all accumulated feature-flag debt when it is not. |
| Release Train | pattern | Deployment Strategies | A fixed-cadence release where whatever is ready by the departure time ships, and whatever is not waits for the next one. |
| Relocate Disposition Seventh R, Hypervisor-Level Migration, Cluster Relocation | pattern | The Six Rs | The migration treatment that moves a hypervisor-managed estate to a hypervisor-compatible target with the application and operating system unchanged, so the unit of work is the cluster rather than the machine … |
| Remote Module Loading | pattern | Module Federation | Importing code from another application's build at runtime, with the versioning, availability and failure questions that a normal import does not have. |
| Replay Pipeline | pattern | Kappa vs Lambda | A single processing path that produces both live and historical results by re-running the same code over retained input, replacing the two-path Lambda arrangement. |
| Replication in Practice Read Replicas, Multi-Primary | pattern | Replication | Copying data across nodes for availability and read capacity, and the lag that turns into user-visible bugs if session guarantees are not designed in. |
| Reported Versus Desired State | pattern | Digital Twin | Holding two representations of a device — what it last said it was, and what it should be — with the gap between them driving reconciliation. |
| Request-Scoped Identity Principal-Bound Response, Per-Request Ownership Check | pattern | Secure API Design | Binding every piece of shared mutable state a response is assembled from to the principal who asked for it, so that a bug in pooling or caching produces an error rather than another customer's data. |
| Reranking Cross-Encoder Reranking | pattern | Reranking | Retrieving a wide candidate set cheaply, then reordering it with a more expensive model that scores each candidate against the query directly. |
| Reservation Ledger Stock Ledger, Reserve-Then-Commit | pattern | Transactions & Isolation | Modelling stock as an append-only sequence of reservations, releases and allocations rather than as a mutable count - so the number has a history, can be reconciled, and does not become a single point of contention. |
| Retrieval Pipeline Stages | pattern | RAG Architecture | The stages that turn a user question into grounded context — query processing, retrieval, reranking and assembly — each independently tunable. |
| Retrieval-Augmented Generation RAG | pattern | AI-Era Architecture | Retrieving relevant documents at query time and putting them in the model's context, so answers are grounded in your data rather than in training data. |
| Retry Budget Retry Ratio Limit, Adaptive Retry Throttling | pattern | Retries & Backoff | A global cap on retries as a proportion of total requests, which bounds retry amplification during broad failures in a way that per-request retry counts cannot. |
| Reverse ETL Operational Analytics, Data Activation | pattern | Reverse ETL | Pushing modelled data from the warehouse back into operational systems so that business tools act on the same definitions analysts report on. |
| Reverse Proxy | pattern | Networking | A server that receives client requests on behalf of one or more backends, forwarding them and returning the response. |
| Reverse Proxy in Practice | pattern | Reverse Proxies | A server that sits in front of origins and terminates, caches, routes and protects — the cheapest place to solve several problems at once. |
| Reverse Replication Cutover Backflow Replication, Bidirectional Cutover Window, Rollback Stream | pattern | Coexistence Patterns | Keeping a replication stream flowing from the new store back to the old one after writes have moved, so that rolling back stays a routing change instead of a restore from backup. |
| Revocation Path Fast-Propagation Channel, Fail-Closed Config | pattern | Cloud Governance | A separate, short-TTL, fail-closed channel for security-critical configuration changes, coexisting with an eventually-consistent cached channel for ordinary configuration. |
| Row-Level Security RLS, Fine-Grained Access Control | pattern | Row & Column-Level Security | Access control enforced by the data platform on which rows a given principal can see, rather than by each application or query. |
| Runtime Composition | pattern | Micro-Frontends | Assembling one page from independently deployed fragments at request or render time, which is what makes micro-frontends more than separate builds. |
| Saga Compensating Transaction, Long-Running Transaction | pattern | Distributed Transactions | A sequence of local transactions across services where failure is handled by compensating actions rather than by rollback. |
| Saga | pattern | Distributed Systems | A sequence of local transactions across services where each step has a compensating action that semantically undoes it if a later step fails. |
Nothing on this page matches. Search the whole glossary.