PreemptionTaxScenario

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.

Forced preemptions of an n_devices job whose checkpoint is checkpoint_bytes.

Notes

Equation: result = PreemptionTaxModel.solve(…) with the declared n_devices, checkpoint_bytes, lost_work inputs.

Assumptions: The inputs instantiate the assumptions documented by mlsysim.solvers.PreemptionTaxModel.solve; accessing result evaluates that solver again.

Excludes: Any extra mechanism beyond PreemptionTaxModel; this record supplies configuration and does not collect performance measurements.

References

Derivation: This configuration forwards its stored inputs to mlsysim.solvers.PreemptionTaxModel.solve. The solver contract records the equations and their provenance.

Examples

The named scenario supplies illustrative inputs; the output is calculated.

>>> from mlsysim import Scenarios
>>> case = Scenarios.ForcedPreemption175BFineTune64
>>> round(float(case.result.lost_work.to('min').magnitude), 3)
15.0