CheckpointStormScenario
Data scenarios · Implementation: mlsysim/engine/scenario_definitions/data.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
Every one of replicas writes workload’s full resumable state through one ingest path.
Notes
Equation: Compose mlsysim.physics.quantities.memory_from_params, mlsysim.solvers.CheckpointStormModel.
Assumptions: The stated fields are inputs to mlsysim.solvers.CheckpointStormModel; its documented validity regime, units, and modeling choices apply unchanged.
Excludes: This scenario adds no measurement, empirical calibration, or mechanisms beyond mlsysim.solvers.CheckpointStormModel.
References
Derivation: this scenario binds its input fields to the referenced implementation contracts: mlsysim.physics.quantities.memory_from_params, mlsysim.solvers.CheckpointStormModel. The methods below identify any additional arithmetic.
Examples
>>> from mlsysim import Scenarios
>>> case = Scenarios.Llama2_70BCheckpointStorm1K
>>> type(case.result).__name__
'CheckpointStormResult'Attributes
| Name | Description |
|---|---|
| bytes_per_param | |
| ingest_bandwidth | |
| replicas | |
| result | Evaluate this scenario through mlsysim.solvers.CheckpointStormModel.solve. Equation: |
| workload |
workload
Type: Workload.
Declared default: required.
bytes_per_param
Type: Quantity.
Declared default: required.
replicas
Type: int.
Declared default: required.
ingest_bandwidth
Type: Quantity.
Declared default: required.
result
engine.scenario_types.CheckpointStormScenario.result
Evaluate this scenario through mlsysim.solvers.CheckpointStormModel.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.