EncryptedInferenceLatencyScenario
Reliability scenarios · Implementation: mlsysim/engine/scenario_definitions/reliability.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
One inference at plaintext_latency, with symmetric encryption around it (encryption_overhead) and computed under homomorphic encryption (fhe_multiplier).
Notes
Equation: Add symmetric transport overhead to baseline latency; multiply baseline latency for the FHE scenario.
Assumptions: The encryption cost and homomorphic multiplier are supplied scenario inputs.
Excludes: Cipher-specific implementation, accuracy effects, and independently predicted cryptographic cost.
References
Derivation: Add symmetric transport overhead to baseline latency; multiply baseline latency for the FHE scenario. Individual property contracts identify the exact accounting expression or reusable calculation. Registry entries supply the scenario inputs; this class does not turn them into measured outcomes.
Examples
Evaluate a named scenario using its declared inputs.
>>> from mlsysim import Scenarios
>>> scenario = Scenarios.HealthMonitoringEncryptedInference
>>> round(float(scenario.encrypted_transport_latency.magnitude), 4)
20.5Attributes
| Name | Description |
|---|---|
| encrypted_transport_latency | Milliseconds for plaintext inference plus the supplied symmetric-encryption overhead. |
| encryption_overhead | |
| fhe_latency | Seconds for plaintext inference scaled by the assumed homomorphic-encryption multiplier. |
| fhe_multiplier | |
| plaintext_latency |
plaintext_latency
Type: Quantity.
Declared default: required.
encryption_overhead
Type: Quantity.
Declared default: required.
fhe_multiplier
Type: float.
Declared default: required.
encrypted_transport_latency
engine.scenario_types.EncryptedInferenceLatencyScenario.encrypted_transport_latency
Milliseconds for plaintext inference plus the supplied symmetric-encryption overhead.
Equation: (plaintext_latency + encryption_overhead).to('ms').
fhe_latency
engine.scenario_types.EncryptedInferenceLatencyScenario.fhe_latency
Seconds for plaintext inference scaled by the assumed homomorphic-encryption multiplier.
Equation: (plaintext_latency * fhe_multiplier).to('second').