Skip to main content

KvAdmissionScenario

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 replica’s KV budget against one context length (KvAdmissionModel).

Notes

Equation: Result = mlsysim.solvers.KvAdmissionModel.solve with the named scenario fields; no extra fitting or postprocessing is applied.

Assumptions: The supplied inputs use the units and validity regime documented by KvAdmissionModel.solve; this class binds one reproducible configuration.

Excludes: Any behavior excluded by KvAdmissionModel.solve, and measurement or calibration of the supplied inputs.

References

Implementation reference: mlsysim.solvers.KvAdmissionModel.solve documents the composed equation and its sources.

Examples

The named registry scenario supplies the stated inputs; this is a modeled calculation, not a new measurement.

>>> from mlsysim import Scenarios
>>> scenario = Scenarios.Llama3_70BLongContextAdmissionDgxH100
>>> scenario.result.head_dim
128

Attributes

Name Description
accelerators
context_tokens
hbm_per_accelerator
kv_bytes_per_elem
requests
reserve
result Return mlsysim.solvers.KvAdmissionModel.solve for this scenario; output fields retain the units and feasibility semantics specified by that solver.
weight_bytes_per_param
workload

Methods

Name Description
context_fit Unrounded per-request context capacity in tokens: remaining KV bytes divided by positive request count and bytes per token. Uses the admission model’s kv_budget and kv_per_token; supply requests or the scenario default.

context_fit

engine.scenario_types.KvAdmissionScenario.context_fit(requests=None)

Unrounded per-request context capacity in tokens: remaining KV bytes divided by positive request count and bytes per token. Uses the admission model’s kv_budget and kv_per_token; supply requests or the scenario default.

Equation: The calculation follows the stated accounting rule; the method body shows each intermediate term.

workload

Type: Workload.

Declared default: required.

accelerators

Type: int.

Declared default: required.

hbm_per_accelerator

Type: Quantity.

Declared default: required.

weight_bytes_per_param

Type: Quantity.

Declared default: required.

reserve

Type: Quantity.

Declared default: required.

context_tokens

Type: int.

Declared default: required.

kv_bytes_per_elem

Type: Quantity.

Declared default: required.

requests

Type: Optional[int].

Declared default: None.

result

engine.scenario_types.KvAdmissionScenario.result

Return mlsysim.solvers.KvAdmissionModel.solve for this scenario; output fields retain the units and feasibility semantics specified by that solver.

Equation: mlsysim.solvers.KvAdmissionModel.solve (scenario fields).

Back to top