TrafficSurgeScenario
Serving scenarios · Implementation: mlsysim/engine/scenario_definitions/serving.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
A service sized for normal_rate that receives surge_rate; latency in each regime.
Notes
Equation: The calculation members below define surge_factor.
Assumptions: normal_rate is positive and both rates count comparable requests.
Excludes: A latency prediction from the surge ratio alone.
References
Derivation: A service sized for normal_rate that receives surge_rate; latency in each regime. Each member records the corresponding algebra or named physics function.
Examples
Evaluate the named configuration and display a modeled result.
>>> from mlsysim import Scenarios
>>> case = Scenarios.RecommenderTrafficSurge
>>> value = case.surge_factor
>>> print(f"{value:.3g}")
10Attributes
| Name | Description |
|---|---|
| normal_latency | |
| normal_rate | |
| surge_factor | Surge arrival rate divided by normal arrival rate, dimensionless. |
| surge_latency | |
| surge_rate |
normal_rate
Type: Quantity.
Declared default: required.
surge_rate
Type: Quantity.
Declared default: required.
normal_latency
Type: Quantity.
Declared default: required.
surge_latency
Type: Quantity.
Declared default: required.
surge_factor
engine.scenario_types.TrafficSurgeScenario.surge_factor
Surge arrival rate divided by normal arrival rate, dimensionless.
Equation: (surge_rate / normal_rate).to('').magnitude.