Glossary

Ansatz

A parameterized circuit structure used by variational algorithms. The ansatz defines the search space available to the optimizer.

Backend

A target execution environment: a physical QPU, simulator, or managed cloud device.

Calibration snapshot

A versioned record of device parameters and quality metrics used by compilation and execution.

Circuit depth

The number of sequential operation layers along a circuit’s critical path, under a declared instruction-counting convention. Depth is not execution duration: gate durations, delays, measurement, and classical feed-forward also matter. Circuits with data-dependent control flow may require path-specific or bounded depth descriptions.

Coherence time

The time scale over which a qubit preserves quantum information before environmental interaction degrades it.

Crosstalk

Unwanted influence of operations on qubits or couplers not intended to be affected.

Error budget

A workload’s declared tolerance for error in a specified output metric, with allocations or bounds for relevant error sources. Bias, variance, gate-error rates, and model error cannot be added indiscriminately; combining them requires compatible quantities and justified assumptions. A service reliability error budget is a separate concept derived from an SLO.

Error correction

Encoding logical quantum information across multiple physical qubits and using syndrome measurements to detect and correct errors.

Error mitigation

Post-processing or experimental techniques that reduce the impact of noise without full fault-tolerant correction.

Logical qubit

An encoded qubit protected by an error-correcting code. A logical qubit usually requires many physical qubits.

Native gate

A gate directly supported by the hardware control system. Other gates must be decomposed into native operations.

QPU

Quantum processing unit. The physical quantum device that executes quantum operations.

Readout error

A measurement error in which the classical result differs from the actual measurement outcome.

Shot

One execution of a quantum circuit or program that yields a classical measurement record. The record may contain multiple registers or mid-circuit outcomes; it is a sample, not a complete description of the quantum state.

Surface code

A leading family of quantum error-correcting codes using local stabilizer measurements on a lattice-like layout.

Transpilation

The process of converting a high-level quantum circuit into an equivalent circuit compatible with a target backend’s gates and topology.

flowchart LR
    Source[Source circuit] --> Transpile[Transpilation]
    Transpile --> Native[Native circuit]
    Native --> Shots[Repeated shots]
    Shots --> Counts[Measurement counts]
    Counts --> Estimate[Estimate with uncertainty]

Control IR

An internal representation that sits between compiled circuits and hardware commands. It should encode timing constraints, channel assignments, calibration references, conditional behavior, acquisition windows, and safety constraints.

Dynamic circuit

A circuit that uses measurement results during execution to influence later operations. Dynamic circuits require classical control paths with bounded latency.

Golden circuit

A small, stable benchmark circuit with well-understood expected behavior. Golden circuits are used as canaries for regressions, not as proof that a platform is generally high quality.

Quality-adjusted throughput

A scheduling metric that values completed jobs by the probability that their results are useful, rather than by raw QPU utilization or raw job count.

Near-time loop

A classical-quantum feedback loop operating on millisecond-to-second timescales, such as choosing the next parameters in a variational algorithm.

Real-time loop

A feedback loop operating on device-control timescales, often nanoseconds to microseconds, such as measurement-conditioned correction.

Session

A scheduling mode for iterative workloads where related jobs are grouped to reduce interruption and improve continuity across a quantum-classical loop.

Trust report

A user-facing report that attaches provenance, calibration state, compiler metadata, mitigation settings, shot count, benchmark context, and uncertainty to a quantum result.

v7 additions

Autonomous calibration: Calibration in which software proposes or executes tuning actions under explicit guardrails, approval modes, and rollback rules.

Capability contract: A structured target description that records native operations, connectivity, dynamic features, quality evidence, limits, policy constraints, and cost assumptions.

Commissioning: The staged release process that turns assembled quantum hardware or a new target into an operational system with validated envelopes, support paths, and evidence retention.

Compliance boundary: The set of access, export-control, data, security, and claims controls governing how quantum resources and technical information can be used or shared.

Evidence package: A structured bundle of problem intent, source, compiled workload, target context, calibration context, execution metadata, raw results, statistical analysis, and review decision.

Federation: A platform capability that routes workloads across multiple QPUs, simulators, providers, or classical resources while preserving constraints and evidence.

Reliability debt: An accepted workaround, waiver, or degraded condition that remains in production and must be tracked until retired.

Signal-path inventory: A documented map of timing references, firmware, waveform generation, cables, control electronics, readout chains, and feedback paths.

runtime contract

A versioned agreement describing program semantics, target requirements, compilation settings, execution settings, resilience policy, and evidence artifacts for a quantum run.

target profile

A declared set of capabilities and limits supported by a quantum target, such as available operations, dynamic-control support, timing constraints, and result formats.

pulse artifact

A governed pulse-level control package containing waveform parameters, safety bounds, target qubits/channels, validation results, and rollout state.

mitigation debt

Operational debt created when error mitigation hides hardware degradation, underreports uncertainty, increases capacity cost, or creates opaque provider dependency.

evidence package

A collection of raw data, compiled programs, target snapshots, calibration snapshots, mitigation metadata, analysis code, and statistical outputs that support a quantum claim.

resource bundle

A scheduled group of CPU, GPU, QPU, storage, and network resources required by a hybrid quantum workload.

algorithm supply chain

The governed path from mathematical algorithm through executable specification, logical resource estimates, compiler assumptions, verification tests, and runtime package.

Additional v9 terms

Assurance level

A classification of how much review, reproducibility, and audit evidence a quantum result requires before it can support a decision or claim.

Control profile

A machine-readable declaration of the dynamic-control capabilities required by a workload or offered by a target, including mid-circuit measurement, reset, branching, timing, and feed-forward constraints.

Experiment warehouse

A queryable system of record for quantum experiments, including runs, artifacts, target snapshots, metrics, costs, claims, and reviews.

Hardware canary

A small representative hardware workload run before a larger workload to detect target drift, provider incidents, or calibration regressions.

Quantum inner loop

The developer loop from hypothesis to simulator validation, resource estimate, small hardware probe, and reviewed evidence.

Workload contract

A structured description of what a quantum workload requires and what evidence it must produce. The contract may describe a circuit, AHS program, annealing model, hybrid workflow, or fault-tolerant estimate.

v10 glossary additions

Benchmark catalog — A governed collection of benchmark specifications, result cards, statistical contracts, and historical evidence.

Claim rollback — The process of marking a quantum result claim provisional, superseded, or withdrawn after an incident or evidence review.

Evidence ladder — A graduated verification model that matches claim strength to required evidence, from basic sanity checks to audit-grade validation.

Reference configuration — A deployable architecture pattern with known scope, controls, evidence expectations, and exit paths.

Runtime reliability budget — The allowed rate of runtime failures, degraded responses, missing evidence bundles, or contract violations before admission or rollout policies change.

Workflow manifest — A versioned declaration of workload purpose, tasks, backends, budgets, artifacts, statistical plan, and stopping rules.

Additional v11 terms

Benchmark contract

A machine-readable description of a benchmark's workload family, inputs, target requirements, quality metrics, cost envelope, baselines, and evidence requirements.

Claim class

A governance category defining the evidence and approval threshold required before a result may support an internal, customer-facing, scientific, regulatory, investor, or public claim.

Compilation certificate

A signed artifact recording source hash, compiler version, target profile, pass sequence, verification checks, limitations, and approval state for a compiled quantum workload.

Cost envelope

A pre-execution budget and stopping-rule package that caps shots, tasks, active time, simulator use, mitigation overhead, retries, and review cost.

Data residency policy

A policy that constrains where quantum source artifacts, compiled artifacts, classical optimizer state, results, telemetry, backups, and support access may reside.

Hazard registry

A maintained record of physical hazards, controls, PPE, training, owners, inspection cadence, and emergency procedures for quantum facilities.

Knowledge graph

A relationship-oriented operational memory linking workloads, source artifacts, compiler versions, target profiles, calibrations, jobs, results, evidence packages, claims, policies, and cost records.

Safety envelope

The validated facility, equipment, software, and procedural conditions under which a quantum system may be operated, serviced, modified, or shut down.

Timing contract

A versioned control-layer agreement describing clock sources, channel alignment, instruction durations, feedback latency, jitter budgets, and timing telemetry expectations.

v12 glossary additions

Execution bill of materials (execution BOM) — An extension of an SBOM that records software, compiler, runtime, target, calibration, control, and interpretation artifacts that shaped a quantum result.

Policy-as-code — The practice of writing admission, execution, publication, and governance rules as versioned, testable, machine-evaluated policy.

Quantum chaos drill — A controlled fault-injection exercise that tests whether a quantum platform reaches a safe state when a provider, metadata, calibration, routing, or evidence component degrades.

Reanalysis contract — A metadata record describing how a preserved quantum result can be reinterpreted after methods, baselines, or calibration knowledge improve.

Energy per valid shot — A sustainability proxy that divides estimated energy by shots that pass validation and remain usable for the intended evidence package.

Capability negotiation — A broker process that matches workload requirements against current provider and target capabilities before route selection.

Assumption ledger — A maintained set of strategic assumptions, leading indicators, confidence ratings, and review triggers for long-horizon quantum planning.

Fleet health

A workload-specific view of a QPU or provider fleet that combines availability, calibration age, error trends, benchmark probes, queue behavior, telemetry completeness, and user impact.

Entropy service

A production service that validates, conditions, allocates, and audits random material derived from a physical or quantum entropy source.

Timing evidence

Metadata and traces that describe clock references, synchronization, pulse schedules, capture windows, firmware versions, and classifier versions for timing-sensitive quantum experiments.

Benchmark marketplace

A governed registry where benchmark definitions, implementations, submissions, evidence packages, results, challenges, and scorecards can be compared across targets.

Shared quantum testbed

A multi-user quantum facility or program that provides access, support, evidence handling, training, and technology-transfer workflows across institutions.

Model-based systems engineering

An engineering practice that maintains machine-usable relationships among requirements, interfaces, capabilities, hazards, tests, and evidence.

Supervised autonomy

A platform operating mode where automated agents can recommend or perform bounded actions under explicit policies, audit trails, human-review thresholds, and rollback plans.

v14 glossary additions

Standard operating environment (SOE) — The approved baseline of quantum toolchains, target classes, policies, evidence schemas, support paths, and release rules used to make quantum work repeatable.

Quantum zero trust — A resource-centric access model in which every quantum run, control action, evidence read, and administrative operation is evaluated against identity, device, workload, target, data, and policy context.

Data sunset — A controlled reduction, archival, or deletion process for quantum artifacts based on claim dependency, contractual retention, regulated status, and reproducibility needs.

Field operating envelope — The validated environmental, safety, connectivity, operator, and calibration conditions under which a field-deployed quantum system may collect evidence.

QNOC — A quantum network operations center responsible for link health, entanglement-service routing, timing state, incidents, maintenance windows, and network evidence.

Agent action envelope — A machine-readable boundary defining what an automated quantum operations agent may do, including cost, target, data, time, approval, and forbidden-action constraints.

Enterprise quantum broker — A platform component that exposes a stable enterprise API while abstracting providers, enforcing policy, recording evidence, and normalizing cost and result metadata.

v15 assurance and governance terms

Assurance case: A structured claim-argument-evidence record that explains why a quantum result, service, or operational control should be trusted.

Bounded claim: A claim constrained by target, time window, method, statistical confidence, and assumptions.

Change advisory record: A machine-readable record describing a calibration, compiler, runtime, policy, or evidence change and its risk, canaries, approvals, and recovery plan.

Evidence freeze: The immediate preservation of workload, target, compiler, runtime, calibration, telemetry, and result artifacts at the start of an incident.

Mission profile: A record defining the decision, acceptance criteria, evidence level, abort criteria, and review path for a high-value quantum workload.

Portability drill: A controlled migration exercise that tests whether a workload can move across providers while preserving semantic intent, evidence, governance, and economics.

Quantum advantage economics: The discipline of comparing a quantum-enabled workflow against baselines using decision value, cost per validated result, evidence level, and kill criteria.

Signal register: A maintained list of external and internal signals that may alter scenario assumptions or portfolio decisions.

v16 governance, evidence, and maintenance terms

Governance API — A machine-readable interface that turns admission, policy, evidence, control mapping, and review decisions into enforceable platform objects.

Calibration trust exchange — A normalized layer that records target health, calibration freshness, evidence completeness, and trust tiers across providers and target classes.

Workload contract — A versioned specification of workload semantics, allowed transformations, execution requirements, evidence obligations, and claim boundaries.

Tolerance envelope — A predeclared statistical or operational boundary used to decide whether observed quantum behavior remains acceptable for a workload or claim.

Operator reliability — The reliability contribution of interfaces, runbooks, alerts, training, handoffs, and human decision support in quantum operations.

Legal hold scope — A structured set of workload, claim, incident, provider job, time-window, and record identifiers used to preserve relevant evidence.

Semantic operations layer — A canonical data model and event fabric that connects workload, target, calibration, compiler, runtime, evidence, cost, and governance records.

Action envelope — A policy-bound definition of what an automated operations agent may do, including allowed actions, forbidden actions, cost limits, scope, approval thresholds, and rollback requirements.

Maintainer review — A recurring review of reference drift, dependency drift, runbook health, evidence completeness, unresolved postmortem actions, scorecard regressions, and retirement candidates.

v17 additions

Abuse case: a concrete way an actor could misuse, manipulate, or degrade a quantum platform or evidence workflow.

Degradation mode: a predefined reduced-capability operating state that preserves safety, evidence, and user trust during partial failure.

Federated evidence room: a controlled cross-organization workspace for sharing scoped, signed, and policy-governed quantum evidence artifacts.

Logical resource broker: a scheduler that allocates logical qubits, error budgets, decoder capacity, magic-state factories, and verification budgets.

Quantum-network trust boundary: a boundary across quantum channels, classical channels, key managers, trusted nodes, timing systems, or evidence systems where security assumptions change.

Shadow decoder: an independent or delayed decoder used to compare against the production decoder and detect divergence.

Witnessable event: a benchmark or execution event whose occurrence, inputs, or outputs can be independently signed or observed.

v18 additions

Quantum SLI — A measurable indicator of quantum service behavior, such as queue latency, calibration freshness, evidence completeness, or confidence-interval width.

Quantum SLO — A scoped objective that defines acceptable behavior for a quantum service class over a measurement window.

Quantum service reliability error budget — The allowed bad-event fraction or count derived from a declared SLO over its measurement window. Statistical error tolerances and spending limits are separate budgets unless the service contract explicitly defines a combined success predicate.

Compliance drift — Divergence between the authorized operating baseline and the actual state of controls, configurations, data, or behavior.

Exercise design packet — A structured plan for a tabletop, simulation, game day, or resilience drill.

Quantum role card — A record that defines an operating role, mission, responsibilities, skills, backup expectations, and recertification cadence.

Quantum service catalog — A governed inventory of service classes, capabilities, constraints, trust levels, cost models, and exit paths.

Quantum data product — A reusable, owned, versioned data asset such as a calibration timeline, evidence manifest set, benchmark corpus, or workload registry.

v19 additions

Readiness gate — A structured review point that decides whether a quantum workload, platform service, or claim may advance to the next maturity level.

Operating readiness level — A maturity assessment of whether a workload is supportable, reproducible, governed, costed, and diagnosable, distinct from pure technology readiness.

Quantum control-plane reference implementation — A minimal set of services for identity, policy, workload intake, routing, evidence capture, eventing, cost attribution, and lifecycle state.

Quantum data contract — A versioned schema and semantic agreement for quantum artifacts such as workload requests, circuit manifests, calibration snapshots, results, and evidence packages.

Confidence ladder — A planned sequence of validation steps that increases trust in a quantum claim while controlling validation cost.

Validation economics — The discipline of matching validation depth to decision stakes, evidence value, and resource cost.

Quantum risk quantification — A risk-management approach that estimates exposure across calibration, compiler, evidence, provider, algorithm, governance, talent, and economic uncertainties.

Interoperability lab — A controlled environment for testing APIs, events, artifacts, runtime behavior, evidence completeness, and provider exit paths.

Queue economics — The study of how scarce QPU access, reservations, shot budgets, drift windows, priority policy, and statistical stopping rules affect platform value.

Industry adoption playbook — A domain-specific operating plan that maps quantum risks, use cases, baselines, constraints, gates, owners, and metrics.

v20 additions

configuration baseline: A versioned bundle of hardware, calibration, compiler, runtime, data, access, and evidence state used to determine whether a quantum result is reconstructable and reviewable.

digital thread: A linked chain of identifiers and artifacts that connects design, fabrication, calibration, execution, analysis, evidence, and claims.

build attestation: A signed or otherwise verifiable statement describing how a software or experiment artifact was produced.

notebook promotion: The process of turning exploratory notebook logic into parameterized scripts, automated workflows, runbooks, and reviewable evidence packages.

statistical process control: A family of monitoring methods that distinguishes expected process variation from signals requiring operational action.

developer portal: A discovery and self-service interface that exposes approved services, templates, docs, ownership, status, and support routes.

disaster domain: A boundary inside which a failure can invalidate service, evidence, safety, or trust assumptions.

docs-like-code: A documentation operating model in which docs are versioned, reviewed, tested, and released using software-delivery practices.

Workload identity

A cryptographically verifiable identity assigned to a service or job rather than a human user. In quantum platforms, workload identity is used to prove which broker, scheduler, connector, or evidence writer acted on a run.

Trace envelope

The metadata that binds a distributed trace to a quantum evidence package, including the root trace ID, critical spans, provider job mapping, and retention rule.

Quantum lineage facet

A small metadata object attached to a run, job, or dataset that records quantum-specific context such as circuit hash, calibration baseline, compiler target, shot count, mitigation profile, and evidence ID.

Signed evidence package

A result bundle whose critical artifacts have hashes, signatures, attestations, and provenance sufficient for later verification.

Progressive quantum rollout

A controlled deployment pattern for compiler, runtime, mitigation, scheduler, provider, or evidence changes, using feature flags, guardrails, blast-radius limits, and rollback plans.

Operator copilot

An AI-assisted interface for quantum operators that retrieves, summarizes, drafts, and recommends actions under explicit tool-use, policy, evidence, and human-review controls.

v22 additions

assurance graph

A graph of workloads, executions, evidence, policies, identities, reviews, and claims used to answer audit and reproducibility questions.

transparency log

An append-only log that records artifact commitments and proofs so evidence can be independently checked for later tampering.

certified component marketplace

A governed catalog of reusable quantum components with versioning, ownership, signatures, evidence, caveats, and certification levels.

compliance package

A controlled case file for regulated or high-risk quantum workloads, including classification, approvals, execution constraints, evidence, and release rules.

quantum advisory

A structured notice about a software, compiler, mitigation, calibration, provider, or component defect that may require quarantine, patching, or revalidation.

retention class

A policy-defined category controlling how long a quantum data object is retained, who can access it, and how it is disposed.

control room

The operational surface that integrates fleet health, workload state, evidence completeness, advisories, costs, and governance actions.

v23 additions

compatibility window

A bounded period during which old and new quantum platform behavior are both supported so workloads, evidence, and claims can be compared before retirement of the old behavior.

replay envelope

A durable artifact that records the workload, compiler, target, calibration, runtime, mitigation, analysis, and retention context needed to reconstruct or replay a quantum result.

time-travel debugging

The ability to inspect and reconstruct what the platform believed was true when a result was accepted, including historical contracts, calibration baselines, and evidence state.

quantum red team

A security-testing team or exercise that targets quantum-specific failure modes such as forged evidence, route downgrades, calibration poisoning, plugin abuse, or claim manipulation.

synthetic quantum load

Generated workload traffic used to test the broker, scheduler, provider adapters, evidence pipeline, reservations, and capacity policies without relying on live user demand.

quantum gateway

The controlled entry point for workload submissions, routing, identity, authorization, traffic shaping, trace propagation, and contract validation.

SLO-as-code

A declarative representation of service-level objectives that can be versioned, reviewed, validated, and connected to alerting and error-budget policy.

portfolio control tower

An executive reporting surface that converts quantum operational data into decisions about funding, risk, claims, capacity, talent, and use-case gates.

quantum utility

A mature quantum platform whose capabilities are discoverable, measurable, safe, costed, auditable, and consumable under explicit operating contracts.

Continuous control probe

A small, repeatable test that checks whether a specific control still works and emits evidence that can be stored, queried, and reviewed.

Quantum runtime contract

A versioned description of runtime API behavior, execution modes, target semantics, mitigation defaults, quota accounting, and evidence obligations.

Quantum entitlement

A permission boundary describing which providers, backends, modes, data classes, and cost centers a workload identity may use.

Quantum telemetry normalization

The process of applying consistent names, units, attributes, trace context, and resource metadata across broker, compiler, runtime, provider, and evidence systems.

Quantum trust portal

A customer-facing interface that publishes curated, signed, and redacted assurance information about platform controls, runtime contracts, data handling, incidents, and evidence models.

Quantum secure software factory

A controlled build, test, signing, release, and promotion path for quantum platform code, SDK wrappers, compiler plugins, runtime extensions, evidence validators, policy bundles, and productionized research workflows.

Additional technical sources: [R241], [R263], [R288], [R296].