Terminology

2240 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 areas2240 Architecture Fundamentals77 Distributed Systems107 Data Architecture110 Cloud Architecture94 Networking88 API & Integration Architecture82 Reliability & Resilience80 Observability75 Performance & Capacity Engineering72 Security Architecture86 Cost Architecture & FinOps71 Business Architecture67 Architecture Communication67 Enterprise Architecture66 Legacy Modernization71 AI-Era Architecture74 Software Architecture & Engineering71 Architecture Patterns71 Architecture Decision-Making63 The Architect's Meta-Skills61 Delivery & Release Engineering66 Platform Engineering & Developer Experience70 Testing & Quality Architecture71 Data Platform Architecture69 Streaming & Real-Time Data69 Data Governance & Semantics74 Frontend & Experience Architecture66 Edge, Mobile & IoT68 Regulatory & Data Protection Architecture65 Assurance, Audit & Model Risk69

Kind Any concept806 practice738 pattern379 metric133 case-study77 tool62 protocol45

74 terms shown.

Term	Kind	Topic	What it is
Aggregation Threshold Policy Minimum Cohort Rule, k-Threshold Output Policy	pattern	Data Sharing & Clean Rooms	A rule that only returns results computed over at least a minimum number of subjects, plus limits on overlapping queries, so that shared analytics cannot be resolved back to an individual.
Airbnb Minerva: One Definition of a Metric Minerva Metrics Platform	case-study	Semantic Layer	Airbnb built a central metrics platform because the same business question was producing different answers depending on which dashboard you opened.
BI Governance	practice	BI Governance	Keeping business intelligence trustworthy — through certified definitions and a semantic layer, without preventing analysts from working.
Business Glossary	tool	Business Glossary	An agreed set of business term definitions, owned by the business and linked to the physical data assets that implement them.
Certified Report	practice	BI Governance	A small set of reports explicitly endorsed as authoritative, with an owner and a review cadence, distinguished from everything else people have built.
Classification Propagation Label Inheritance, Lineage-Aware Classification, Derived Sensitivity	practice	Sensitivity Labelling	Carrying a column's sensitivity classification through every derived table, view and export, so that copying restricted data into an aggregate does not silently strip its protection.
Classification Propagation Labels Travelling Through Lineage, Derived-Data Sensitivity	practice	Sensitivity Labelling	Carrying a sensitivity label through every derivation, so a restricted column copied into an aggregate, a training set or an export remains subject to its controls.
Code List Governance	practice	Reference Data	Managing the small shared enumerations — currencies, country codes, statuses, product categories — whose uncontrolled change breaks systems quietly.
Computation-Only Collaboration	pattern	Data Sharing & Clean Rooms	Two parties analysing their combined data without either receiving the other's records, by permitting only queries whose outputs are aggregate.
Contract Enforcement Point Producer-Side Enforcement, Shift-Left Contract Checking	practice	Data Contracts	Where a data contract is actually checked - which determines whether a breaking change fails in the producer's build or in a consumer's pipeline at 2 a.m.
Data Access Model	concept	Data Access Models	The scheme by which permission to data is granted, whether by role, by attribute, by purpose, or by request with expiry.
Data Catalogue	tool	Data Catalog	A searchable inventory of data assets with their schemas, owners, lineage, quality signals and usage.
Data Clean Room	pattern	Data Sharing & Clean Rooms	A controlled environment where two parties analyse their combined data without either gaining access to the other's raw records.
Data Contract Producer Contract, Schema Agreement, Published Interface	pattern	Data Contracts	An explicit agreement between a data producer and its consumers covering schema, semantics, quality and change policy - which converts an implicit dependency on someone else's internal table into a managed int…
Data Contract	pattern	Data Contracts	An explicit, versioned, enforceable agreement between a data producer and its consumers covering schema, semantics, quality and delivery.
Data Mesh Federated Governance	concept	Data Governance & Semantics	Central agreement on the rules that must be global, with domain autonomy over everything else.
Data Mesh in Practice	concept	Data Mesh	Decentralising data ownership to the domains that produce it, with a self-serve platform and federated governance — an organisational change more than a technical one.
Data Observability	practice	Data Observability	Continuous automated monitoring of data health — freshness, volume, schema, distribution and lineage — designed to detect silent data failures.
Data Product Published Dataset, Owned Data Asset, Analytical Interface	concept	Data Products	A dataset published deliberately as an interface with a named owner, documented semantics, quality guarantees and a change policy - as distinct from a table that consumers happen to be able to reach.
Data Product	concept	Data Products	A curated, documented, owned dataset treated as a product with consumers, a service level and a lifecycle, rather than as the output of a pipeline.
Data Quality Dimensions	concept	Data Quality Dimensions	The standard axes along which data quality is measured, which turn a vague complaint into a testable assertion.
Data Steward	concept	Data Stewardship	The named individual accountable for the definition, quality and appropriate use of a specific data domain.
Dataset Deprecation Policy Dataset Sunset Policy, Retirement Notice Period	practice	Data Products	The written rule stating how a published dataset or column is retired - the notice period, how consumers are identified, and what happens if they do not move - without which nothing can ever be removed.
Decision Rights Over Data	concept	Data Governance & Semantics	Who is entitled to decide what a data element means, who may see it and how long it is kept — the substance of governance once the policy documents are set aside.
Deletion Propagation Purge Fanout, Erasure Propagation	pattern	Retention & Purge	The mechanism that carries a deletion from the primary store to every derived copy - which is where retention and erasure obligations are actually met or missed.
Deletion Verification	practice	Retention & Purge	Proving that a deletion actually removed the data everywhere it existed, which is a harder problem than issuing the delete.
Deterministic Tokenisation	practice	Tokenisation & Masking	Replacing a sensitive value with a stable surrogate so the same input always yields the same token, preserving joins while removing the value.
Differencing Attack Query Differencing, Reconstruction Attack	concept	Data Sharing & Clean Rooms	Inferring an individual's data from the difference between two permitted aggregate results - the reason that returning only aggregates is not by itself a privacy guarantee.
Domain Data Ownership	concept	Data Mesh	Making the team that generates data accountable for publishing it fit for others to use, rather than handing raw extracts to a central team.
Domain Publishing Readiness Data Product Admission Bar, Domain Onboarding Gate	practice	Data Mesh	The small set of capabilities a domain team must demonstrably have before it is allowed to publish a data product, which is what separates decentralised ownership from renaming existing teams as domains.
Dynamic Data Masking Query-Time Masking, Policy-Applied Redaction	pattern	Tokenisation & Masking	Redacting a column's value as the query returns it, based on who is asking - so one physical table serves cleared and uncleared audiences, while every object derived from it escapes the rule.
Erasure Amplification Deletion Write Multiplier, Purge Fan-Out	concept	Retention & Purge	The ratio between the bytes rewritten to honour a deletion request and the bytes actually belonging to the subject, which decides whether on-demand erasure is feasible or must be batched.
Glossary Term Half-Life Definition Decay Rate, Term Attestation Age	metric	Business Glossary	The time after which half a glossary's definitions no longer match the SQL that implements them, measured by sampling rather than assumed, and used to set the review budget a glossary can actually afford.
Golden Record	concept	Master Data Management	The single authoritative version of an entity assembled from conflicting sources, and the survivorship rules that decide which value wins.
Governed Metric Set The Twenty Numbers, Governance Boundary	practice	Self-Service vs Governed	The deliberately small collection of metrics defined once with an owner, with everything else left explicitly self-service - because governing all analysis becomes a bottleneck and is abandoned.
Governed Path Bypass Shadow Access Route, Route-Around	concept	Self-Service vs Governed	Any route to data that reaches it without passing the controls the organisation believes are in force, which is what makes a correct policy optional rather than binding.
Governed Self-Service	concept	Self-Service vs Governed	Letting anyone analyse anything while constraining what may be treated as authoritative, which resolves the apparent conflict between speed and consistency.
Identifier Reuse Hazard Code Recycling, Key Reassignment	concept	Reference Data	The silent corruption that occurs when a retired code or key is later assigned to a different real-world entity, so historical facts join to the wrong meaning with no error anywhere.
Just-in-Time Data Access Time-Boxed Data Grant, Expiring Analytical Access	pattern	Data Access Models	Replacing permanent read grants with short-lived grants issued on request with a recorded purpose, so that entitlement decays by default instead of accumulating for as long as someone stays employed.
Label Inheritance Classification Propagation, Sensitivity Inheritance	practice	Sensitivity Labelling	Automatically assigning a derived dataset the highest sensitivity of its inputs, so that aggregates and transformations do not silently escape the controls that applied to their sources.
Lineage Coverage Harvest Coverage, Lineage Completeness Ratio	metric	Data Catalog	The share of known data assets whose upstream and downstream edges were successfully derived on the latest harvest, which is the only signal that separates an asset with no dependencies from one the parser cou…
Master Data Management MDM, Golden Record	practice	Master Data Management	Establishing a single authoritative version of core business entities that appear inconsistently across many systems.
Match Confidence Band Match Threshold Band, Review Band	metric	Master Data Management	The score range between automatic merge and automatic reject in record matching, which is where a consolidation programme's permanent staffing cost lives and where its worst errors are prevented.
Match Rate Bias Matched-Cohort Bias, Identity Coverage Bias	concept	Data Sharing & Clean Rooms	The systematic skew introduced when only the identifiable subset of each party's population joins, so a shared measurement describes the matched cohort rather than the market it is quoted about.
Metadata Harvesting	practice	Data Catalog	Populating a catalogue automatically from the systems that hold the data, because anything requiring manual entry will be incomplete within a quarter.
Metric Certification Certified Metrics, Trusted Tier	practice	BI Governance	Marking a small set of metrics and datasets as governed and authoritative, so that exploration stays free while anything asserted about the business comes from a controlled definition.
Metric Definition Layer	pattern	Semantic Layer	Business metrics defined once in a versioned, tested place and served to every consuming tool, so the number cannot differ by which tool asked.
Over-Classification Label Inflation, Maximum Propagation Collapse	concept	Sensitivity Labelling	The state in which so much of an estate carries the highest sensitivity label that its controls become unworkable, and people route around them - producing less protection than a coarser scheme would.
Partition Immutability Promise Closed Partition Guarantee, Correction-Row Policy	pattern	Data Products	A published commitment that once a time partition is closed its rows will never change, with corrections issued as new rows - so consumers can checkpoint and reconcile instead of reloading history.
Pipeline Anomaly Detection	practice	Data Observability	Monitoring row counts, distributions, freshness and schema for unexplained change, because data pipelines fail silently far more often than they error.
Point-in-Time Correctness As-Of Join, Temporal Correctness, Leakage Prevention	concept	Semantic Layer	Retrieving each feature's value as it was at the moment of the event rather than its current value, which is what prevents a model from training on information that did not exist yet.
Policy Bypass Surface Enforcement Gap, Derived-Object Leak	concept	Row & Column-Level Security	The set of objects and paths through which data protected by a row or column policy can be read without that policy being evaluated - typically derived objects refreshed under a privileged identity.
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.
Producer Obligation	concept	Data Contracts	What a data contract binds the producing team to do, and — crucially — what happens in their pipeline when they are about to break it.
Producer-Enforced Contract Contract That Fails the Build, Upstream Enforcement	practice	Data Contracts	A data contract checked in the producing team's own pipeline, so a breaking change is blocked before it ships - without which the contract is documentation and the breakage is discovered downstream.
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.
Quality Dimension Threshold	metric	Data Quality Dimensions	The stated numeric level at which a dataset is fit for its purpose on a given quality dimension, plus what happens when it is not met.
Reference Data	concept	Reference Data	The slowly-changing shared vocabularies — country codes, currencies, product categories, tax rates — whose inconsistency causes disproportionate trouble.
Retention and Purge	practice	Retention & Purge	Deleting data when its lawful or business purpose has ended, and being able to prove that it happened.
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.