Skip to main content

DominantShareScenario

Optimization scenarios · Implementation: mlsysim/engine/scenario_definitions/optimization.py.

Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.

Dominant-resource fairness by progressive filling of indivisible bundles on capacity.

Resources are accelerators, CPU cores, memory (GB), and network (Gb/s). Each tenant’s dominant share is its largest share of any one resource; progressive filling grants the next bundle to the tenant with the lowest dominant share until no bundle fits.

Notes

Equation: Compose mlsysim.labs.experiments.v2_08_experiments.dominant_resource_fairness.

Assumptions: The stated fields are inputs to mlsysim.labs.experiments.v2_08_experiments.dominant_resource_fairness; its documented validity regime, units, and modeling choices apply unchanged.

Excludes: This scenario adds no measurement, empirical calibration, or mechanisms beyond mlsysim.labs.experiments.v2_08_experiments.dominant_resource_fairness.

References

Derivation: this scenario binds its input fields to the referenced implementation contracts: mlsysim.labs.experiments.v2_08_experiments.dominant_resource_fairness. The methods below identify any additional arithmetic.

Examples

>>> from mlsysim import Scenarios
>>> case = Scenarios.DrfTwoTenantExample
>>> type(case.result).__name__
'dict'

Attributes

Name Description
bundles
capacity
result Evaluate this scenario through

capacity

Type: Dict[str, float].

Declared default: required.

bundles

Type: Dict[str, Dict[str, float]].

Declared default: required.

result

engine.scenario_types.DominantShareScenario.result

Evaluate this scenario through mlsysim.labs.experiments.v2_08_experiments.dominant_resource_fairness. Equation: pass the stated fields to that implementation; its contract defines the calculation and feasibility tests. Units: each returned field follows the referenced result schema; no independent equation is evaluated by this wrapper.

Back to top