Layer F: Miscellaneous simulation infrastructure services.

1. Services Configuration

services:
  # Data Recording
  recording:
    enabled: true
    format: HDF5
    path: "output/x15_run_{timestamp}.h5"
    signals: ["X15.*", "Environment.*"]  # Glob patterns
    rate_hz: 100
 
  # Live Telemetry (to Hermes/Daedalus)
  telemetry:
    enabled: true
    protocol: UDP
    port: 5000
    rate_hz: 60
    signals: ["X15.EOM.*", "X15.Aero.mach"]
 
  # Logging
  logging:
    level: INFO
    file: "logs/sim.log"
 
  # Custom Signals (computed at runtime, not owned by any component)
  custom_signals:
    - name: "Derived.TotalEnergy"
      expression: "0.5 * X15.EOM.mass * X15.EOM.velocity^2 + X15.EOM.mass * 9.81 * X15.EOM.position.z"
 
  # External Bindings
  bindings:
    python:
      enabled: true
      module: "pybind11"
    matlab:
      enabled: false
    zmq:
      enabled: true
      port: 5555

2. Data Dictionary & Tooling Integration

The Data Dictionary (auto-generated at Provision, see 02_component_protocol.md) is the single source of truth for all signals. Tooling reads the dictionary—no separate schema file needed.

Tooling Uses

1. Config Validation

# Validate config against a provisioned simulation's data dictionary
$ icarus provision scenarios/x15_mission.yaml --output-dict data_dictionary.yaml
$ icarus validate scenarios/x15_mission.yaml --dict data_dictionary.yaml
 
✓ All signal references valid
⚠ Warning: "X15.Aero.alpah" looks like typo for "X15.Aero.alpha"
✗ Error: "X15.Thruster.force" not found in data dictionary

2. IDE Autocomplete

# VS Code extension reads data_dictionary.yaml for autocomplete
components:
  - type: JetEngine
    config:
      input_altitude: "Env|"  # Suggests: Environment.Atmosphere.Altitude

3. Telemetry/Recording Metadata

// Recorder reads data dictionary for units and descriptions
Recording rec("output.icarec");
rec.LoadDataDictionary("output/data_dictionary.yaml");
rec.AddSignal("X15.Nav.position");  // Dictionary provides: units=m, type=Vec3

4. UI Signal Browser (Daedalus)

Daedalus reads the data dictionary to populate its signal browser with types, units, and descriptions.

3. Why Auto-Generated?

Manual Schema	Auto-Generated Dictionary
Can drift from reality	Always matches actual signals
Extra maintenance burden	Zero maintenance—just run Provision
May have typos	Generated from code—no typos
Incomplete coverage	Complete by definition

Remarks: Workflow: Run icarus provision once to generate the dictionary. Use it for all tooling. Regenerate whenever components change.

4. Custom Signals

Custom signals are computed expressions that don't belong to any component:

custom_signals:
  - name: "Derived.TotalEnergy"
    expression: "0.5 * X15.EOM.mass * X15.EOM.velocity^2 + X15.EOM.mass * 9.81 * X15.EOM.position.z"
    units: "J"
    description: "Total mechanical energy"
 
  - name: "Derived.DynamicPressure"
    expression: "0.5 * Environment.Atmosphere.Density * X15.EOM.velocity^2"
    units: "Pa"
    description: "Dynamic pressure"

These are evaluated by a built-in DerivedSignals component during PostStep.

5. Telemetry Configuration

telemetry:
  enabled: true
  protocol: UDP          # UDP, TCP, ZMQ
  host: "localhost"
  port: 5000
  rate_hz: 60
 
  # Signal selection
  signals:
    - "X15.EOM.*"        # All EOM signals
    - "X15.Aero.alpha"   # Specific signal
    - "!X15.Debug.*"     # Exclude debug signals
 
  # Compression (for high-bandwidth scenarios)
  compression:
    enabled: false
    algorithm: LZ4

6. Logging Configuration

logging:
  level: INFO            # TRACE, DEBUG, INFO, WARNING, ERROR
 
  outputs:
    - type: console
      level: INFO
      format: "[{time}] [{level}] {message}"
 
    - type: file
      path: "logs/sim_{timestamp}.log"
      level: DEBUG
      rotate: true
      max_size_mb: 100
 
    - type: structured
      path: "logs/sim_{timestamp}.jsonl"
      level: TRACE