Entropy and Benchmark Evidence Templates — Operating Quantum Computers

This appendix provides starter templates for quantum randomness services and benchmark marketplace submissions.

Diagram loads as you read

Entropy and Benchmark Evidence Templates · Figure 1

View diagram source

flowchart TB
    Templates[Evidence templates] --> Entropy[Entropy service]
    Templates --> Benchmark[Benchmark marketplace]
    Templates --> Challenge[Challenge process]

Entropy source profile

Illustrative listing · yaml

entropy_source_profile:
  profile_id: quantum_source_profile_v3
  source_type: quantum
  physical_principle: measurement_outcome_variability
  device_id: entropy_device_001
  validation_standard: nist_sp_800_90b_profile
  raw_sample_retention: restricted
  conditioning_required: true
  health_tests:
    startup:
      - repetition_count
      - adaptive_proportion
    continuous:
      - repetition_count
      - adaptive_proportion
      - source_specific_failure_tests
    operational_monitors:
      - throughput_monitor

Diagram loads as you read

Entropy source profile · Figure 2

View diagram source

flowchart LR
    Source[Source profile] --> Tests[Health tests]
    Tests --> Extractor[Extractor]
    Extractor --> Claim[Entropy claim]

Apply health tests to raw digitized noise-source samples before conditioning. Establish a conservative min-entropy estimate, source-specific test thresholds, and a failure response that stops delivery of suspect output. Passing health tests does not establish unpredictability by itself; the source model, entropy assessment, and conditioning design also require validation. NIST SP 800-90B.

Entropy allocation receipt

Illustrative listing · yaml

entropy_allocation_receipt:
  receipt_id: er_2026_04_19_001
  tenant: payments-prod
  purpose: key_generation
  bytes: 4096
  block_id: entropyblk_2026_04_19_001
  source_profile: quantum_source_profile_v3
  extractor_version: extractor_2.4.1
  generated_at: 2026-04-19T12:00:00Z
  delivered_at: 2026-04-19T12:00:02Z
  signature: <signature>

Entropy incident template

Illustrative listing · markdown

# Randomness Service Incident

Incident ID:
Detected at:
Health state:
Source profile:
Affected tenants:
Affected block IDs:

## Detection

## Impact

## Containment

## Revocation or notification decision

## Corrective actions

Diagram loads as you read

Entropy incident template · Figure 3

View diagram source

stateDiagram-v2
    [*] --> Detected
    Detected --> Contained
    Contained --> ImpactAssessed
    ImpactAssessed --> Notified
    ImpactAssessed --> Closed
    Notified --> Closed

Benchmark submission package

Illustrative listing · yaml

benchmark_submission:
  benchmark_id: app_benchmark_001
  benchmark_version: 1.0.0
  submitter: org.example
  workload_family: optimization
  problem_instances:
    - instance_small.json
    - instance_medium.json
  reference_classical_baseline: baseline_v2
  quantum_implementation:
    framework: qiskit
    version: 2.x
  evidence_required:
    - compiled_circuit
    - target_descriptor
    - runtime_options
    - raw_counts
    - post_processing_code
    - cost_estimate

Diagram loads as you read

Benchmark submission package · Figure 4

View diagram source

erDiagram
    BENCHMARK ||--o{ INSTANCE : contains
    BENCHMARK ||--o{ SUBMISSION : receives
    SUBMISSION ||--o{ RESULT : produces
    RESULT ||--o{ EVIDENCE_ARTIFACT : supports

Benchmark result review checklist

Is the benchmark definition versioned?
Is the target descriptor complete?
Are compiler settings disclosed?
Are mitigation settings disclosed?
Is there a classical baseline?
Is uncertainty reported?
Is cost reported?
Are raw counts or sufficient aggregates preserved?
Is the evidence package immutable?
Is the claim wording bounded?

Diagram loads as you read

Benchmark result review checklist · Figure 5

View diagram source

flowchart TB
    Review[Result review] --> Version[Version]
    Review --> Target[Target context]
    Review --> Baseline[Baseline]
    Review --> Uncertainty[Uncertainty]
    Review --> Cost[Cost]
    Review --> Evidence[Evidence]

Challenge template

Illustrative listing · yaml

benchmark_challenge:
  challenge_id: bc_2026_04_19_001
  challenged_result: result_123
  reason: reproducibility_failure
  claimed_issue:
    - missing mitigation settings
    - inconsistent baseline
  proposed_remedy: rerun_with_disclosed_settings
  submitted_by: reviewer@example.org
  status: open

Diagram loads as you read

Challenge template · Figure 6

View diagram source

flowchart LR
    Challenge[Challenge] --> Evidence[Evidence request]
    Evidence --> Rerun[Rerun or audit]
    Rerun --> Decision[Decision]
    Decision --> Publish[Publish annotation]