RingTraceScenario
Communication scenarios · Implementation: mlsysim/engine/scenario_definitions/communication.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
A ring AllReduce traced step by step on one short integer vector per GPU.
Notes
Equation: Compose mlsysim.solvers.RingTraceModel.
Assumptions: One equal-shaped integer vector is supplied per participant and the trace follows the solver’s ring schedule.
Excludes: The trace illustrates data movement; it does not predict kernel or network duration.
References
Derivation: this scenario binds its input fields to the referenced implementation contracts: mlsysim.solvers.RingTraceModel. The methods below identify any additional arithmetic.
Examples
>>> from mlsysim import Scenarios
>>> case = Scenarios.RingTraceFourGpu
>>> round(case.n_devices, 4)
4Attributes
| Name | Description |
|---|---|
| n_devices | Participants represented by the supplied vectors. Equation: len(vectors). Units: |
| result | Evaluate this scenario through mlsysim.solvers.RingTraceModel.solve. Equation: pass |
| vectors |
vectors
Type: List[List[int]].
Declared default: required.
n_devices
engine.scenario_types.RingTraceScenario.n_devices
Participants represented by the supplied vectors. Equation: len(vectors). Units: devices; one vector is supplied per participant.
result
engine.scenario_types.RingTraceScenario.result
Evaluate this scenario through mlsysim.solvers.RingTraceModel.solve. 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.