Portfolio ↗

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

69 terms shown.

Term Kind Topic What it is
Absence-As-Delete Sync Full-Sync Delete Semantics, Destructive Sync concept Reverse ETL A synchronisation mode that treats a record's absence from the source query as an instruction to remove or clear it downstream - so any query returning fewer rows becomes a mass deletion in a live operational …
Airflow: A Scheduler Born From Pipeline Sprawl Apache Airflow case-study Workflow Schedulers Airbnb built Airflow because cron cannot express dependencies, and a data platform's failures are mostly dependency failures.
Analytical Estate Topology concept Data Platform Architecture The full picture of where analytical data lands, is transformed and is served, including the paths that bypass the intended one.
Analytics Cost Control practice Analytics Cost Control The set of design and operational choices that determine whether an elastic data platform costs a predictable amount or an alarming one.
Attribute History Strategy concept Slowly Changing Dimensions The per-attribute decision about whether history is overwritten, versioned or kept alongside the current value, which is a business question rather than a technical one.
Backfill practice Batch Orchestration Reprocessing historical periods through a pipeline after fixing a defect or adding a field, at a scale the pipeline was not sized for.
Batch Orchestration Workflow Orchestration, DAG Scheduling concept Batch Orchestration Running interdependent data jobs in the right order with retries, backfills and failure handling, expressed as a dependency graph rather than a schedule.
Bronze Layer Contract concept Medallion Architecture What the raw landing zone promises and refuses to promise — a faithful replayable copy of the source, with no correction applied.
Business Key Collision Hub Key Collision, False Identity Match concept Data Vault Modelling Two different real-world entities merging into one hub row because the key chosen for it is unique only inside one source system - a corruption that raises no error and is expensive to unpick.
Bytes Scanned Scan Volume, Data Processed concept Analytics Cost Control The volume of data a query reads - which in consumption-priced analytical systems is both the performance limit and the bill, making layout a cost control.
CDC Pipeline Design practice CDC Pipeline Design The concerns that turn log-based change capture from a demo into something that can be relied on — snapshots, ordering, schema change and replay.
Change Volume Ratio Update-to-Insert Ratio, Mutation Rate metric CDC Pipeline Design The proportion of a change stream that is updates and deletes rather than inserts, which decides target storage design, merge cost and whether a table can be append-only.
Clustering Decay Sort Order Drift, Layout Degradation concept Storage Layout & Partitioning The gradual loss of data-skipping as newly written files overlap the sort order of existing ones - so query cost rises month after month with no change to the query or the data volume.
Compaction File Consolidation, OPTIMIZE, Small File Remediation practice File Formats & Compaction Rewriting many small data files into fewer larger ones, with sorting, as a required background process - because streaming ingestion produces small files continuously and a table without compaction degrades si…
Compute Isolation Boundary concept Data Platform Tenancy The line across which one domain's analytical workload cannot affect another's performance, cost attribution or access.
Conformed Dimension Shared Dimension, Conformed Attribute pattern Dimensional Modelling A dimension defined once and used identically by every fact table, so that measures from different processes can be filtered and grouped together without a reconciliation argument.
DAG Dependency concept Batch Orchestration The declared edge stating that one task must not start until another has succeeded, and the difference between that and merely running later.
Data Layout Partitioning and Clustering, File Organisation, Pruning Effectiveness concept Storage Layout & Partitioning How table data is physically organised into partitions and files, which determines how much can be skipped at query time - and is the dominant factor in analytical cost and performance, ahead of cluster size.
Data Pipeline SLA Freshness SLA, Data SLO metric Data Platform Architecture A published commitment about when data will be available and how fresh it will be, turning a pipeline into something consumers can design against.
Data Platform Architecture Analytical Platform Architecture, Data Stack Architecture concept Data Platform Architecture The arrangement of ingestion, storage, transformation, serving and governance that decides whether analytical data arrives on time, means the same thing to everyone and costs a predictable amount.
Data Platform Tenancy concept Data Platform Tenancy How a shared data platform separates domains and customers across storage, compute, catalogue and access control.
Data Vault Modelling pattern Data Vault Modelling A modelling approach separating business keys, relationships and descriptive attributes into hubs, links and satellites, optimised for auditability and change absorption.
Data Virtualisation Query Federation, Logical Data Warehouse pattern Data Virtualisation Querying data where it lives across multiple systems through a single logical layer, without physically consolidating it.
Dataset Ownership Boundary Data Product Boundary, Published Dataset Surface practice Data Platform Architecture The line between a producing team's internal data and the surface it publishes - which is what makes decentralised data ownership safe rather than fragmenting.
Dimensional Model Star Schema, Kimball Model pattern Dimensional Modelling Organising analytical data as fact tables of measurements surrounded by dimension tables of descriptive context.
Fact Grain Multiplication Join Fan-Out, Dimension Row Explosion concept Slowly Changing Dimensions Silent duplication of fact rows when a fact is joined to a versioned dimension without an effective-date predicate, inflating every additive measure in proportion to how often dimension members change.
Grain Declaration practice Dimensional Modelling Stating exactly what one row of a fact table represents, before any column is chosen, because every later decision depends on it.
Hub and Satellite pattern Data Vault Modelling Separating stable business keys from their changing attributes and from their relationships, so each can be loaded independently and kept forever.
Ingestion Pattern concept Ingestion Patterns The mechanism by which data leaves a source system, which determines freshness, source load and how faithfully change is captured.
Lakehouse concept Warehouse, Lake & Lakehouse An architecture combining a lake's cheap open storage with a warehouse's transactions, schema enforcement and query performance, through a table format layer.
Lakehouse Table Format tool Open Table Formats A metadata layer over object storage that adds atomic commits, snapshots and schema evolution to files that otherwise have none.
Layer Contract Guarantee Not Position, Medallion Discipline concept Medallion Architecture The rule that a data layer is defined by the guarantee it makes to consumers, not by its position in a naming convention - without which the layers convey nothing.
Legitimate Divergence Expected Parallel-Run Difference, Correct-on-Both-Sides Delta concept Warehouse Migration The class of differences between an old and a new analytical platform that are correct on both sides, which a parallel run must classify and budget for rather than investigate as defects.
Log-Based Ingestion pattern CDC Pipeline Design Building the pipeline around a database's own change log — an initial snapshot followed by a continuous delta stream, with the two stitched together.
Logical Date Partitioning Execution Date, Idempotent Reprocessing practice Batch Orchestration Making every batch task a pure function of a logical date, so that any run can be repeated safely and a backfill is a range of independent executions.
Medallion Architecture Bronze Silver Gold, Multi-Hop Architecture pattern Medallion Architecture Layering a data platform into raw, cleaned and business-ready zones so that reprocessing is always possible and quality improves in defined steps.
Merge Upsert pattern Ingestion Patterns Applying a batch of changes to a target by matching on a key and inserting, updating or deleting per row, which is expensive and frequently avoidable.
Metadata Read Amplification Planning-Time Amplification, Manifest Read Blow-Up concept Open Table Formats The number of small sequential metadata requests a table format issues before reading any data, which grows with file history and turns a modest storage slowdown into a large query slowdown.
Metadata-Only Migration In-Place Table Adoption, Manifest Generation Migration pattern Open Table Formats Adopting a table format over data files that already exist by generating its metadata in place, so a multi-petabyte conversion becomes a minutes-long commit instead of a full rewrite.
Model Graph Pruning Removing Unused Models, DAG Hygiene practice Transformation Frameworks Periodically removing transformation models with no downstream consumers - the cheapest available reduction in a build's cost and duration, and the one nobody owns.
Open Table Format Iceberg, Delta Lake, Hudi tool Open Table Formats A metadata layer over object storage files that provides atomic commits, schema evolution and time travel, turning a directory of files into a real table.
Operational Data Push pattern Reverse ETL Sending modelled analytical data back into operational tools, which turns a warehouse table into a production dependency with none of the guarantees.
Partition Pruning concept Storage Layout & Partitioning The query engine skipping entire partitions whose values cannot match the filter, which is where most large-table query performance comes from.
Predicate Pushdown Filter Pushdown, Pushdown Optimisation concept Data Virtualisation Sending a filter to the layer that holds the data rather than fetching rows and filtering afterwards, which is the difference between reading megabytes and reading a table.
Query Cost Ceiling Scan Limit, Per-Query Byte Cap pattern Analytics Cost Control A pre-execution limit on the bytes or credits a single query may consume - converting an unbounded bill into a failed query the author can see and fix.
Query Federation pattern Data Virtualisation Executing a single query across several underlying systems without moving the data, and the performance ceiling that pushdown determines.
Report Migration Inventory practice Warehouse Migration The enumerated list of every report, extract and downstream consumer of the legacy warehouse, with usage evidence, which is what makes the migration finite.
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.
Row Group Sizing Row Group Tuning, Parquet Block Size practice File Formats & Compaction Choosing how many rows a columnar file groups into one statistics-bearing unit, which sets both the finest granularity a query can skip and the memory a writer and reader need.
Scanned Bytes metric Analytics Cost Control The volume a query reads, which is what most analytical engines bill for and what almost every optimisation ultimately reduces.
Scheduler Control Plane concept Workflow Schedulers The orchestrator's own state and availability, which becomes a critical dependency for every pipeline it runs.
Silent Coercion Type Coercion on Ingest, Quiet Data Corruption concept Ingestion Patterns An ingestion pipeline converting a value it did not expect rather than rejecting it - the worst available failure mode, because the load succeeds and the data is wrong downstream with nothing indicating it.
Slowly Changing Dimension SCD, Type 2 Dimension pattern Slowly Changing Dimensions A modelling technique for handling attributes that change over time, where the choice determines whether historical reporting stays correct.
Small File Problem concept File Formats & Compaction The severe query degradation caused by a table stored as very many tiny files, where per-file overhead dominates actual data reading.
Snapshot Expiry Snapshot Retention, Metadata Expiry practice Open Table Formats The scheduled removal of a table's old snapshots and the data files only those snapshots referenced, which is what stops time travel from turning every rewritten row into permanent storage.
Snapshot-Stream Convergence Backfill Plus Change Stream, Initial Load Ordering pattern CDC Pipeline Design Starting the change stream before taking the snapshot, so that changes occurring during the backfill are captured rather than silently lost.
Source Load Budget Upstream Read Quota, Analytical Load Cap pattern Data Virtualisation An explicit cap on the concurrency, rows and duration that analytical reads may consume on an operational source - so a dashboard query cannot become a production incident.
SQL Transformation Model concept Transformation Frameworks A named, versioned, tested SELECT statement that declares its own dependencies, turning transformation logic into reviewable software.
Surrogate Key Binding As-Of Dimension Lookup, Version-at-Event Resolution pattern Slowly Changing Dimensions Resolving which version of a dimension row a fact belongs to at load time and storing that version's key on the fact - which is what freezes a historical report instead of letting it drift.
Table Maintenance Budget Maintenance Capacity Reservation, Housekeeping Window practice Workload Isolation Treating compaction, clustering and snapshot expiry as a workload class with its own reserved compute and a guaranteed window, rather than as low-priority work that loses to every deadline.