CompressedReadScenario
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.
Compressed shards read at read_bandwidth, decoded on CPU cores at each codec’s output rate.
Notes
Equation: Use the dimensional products, ratios, or selection rules documented on the calculated properties below.
Assumptions: Compression ratios are uncompressed/compressed sizes; decoder rates are per-core uncompressed output rates and scale linearly across cores.
Excludes: Codec startup, load imbalance, memory contention, and quality changes are absent.
References
Derivation: the calculated properties below state their defining arithmetic and units. They are scenario accounting or input-selection rules, not independently measured relationships.
Examples
>>> from mlsysim import Scenarios
>>> case = Scenarios.NodeStripeCodecs
>>> type(case.output_bandwidth(next(iter(case.compression_ratios)))).__name__
'Quantity'Attributes
| Name | Description |
|---|---|
| compression_ratios | |
| decode_rates | |
| read_bandwidth |
Methods
| Name | Description |
|---|---|
| decode_cores | Ideal CPU-equivalent core demand to keep up with the decompressed stream. Equation: |
| output_bandwidth | Uncompressed byte rate generated by reading a named codec’s compressed data. Equation: |
decode_cores
engine.scenario_types.CompressedReadScenario.decode_cores(codec)Ideal CPU-equivalent core demand to keep up with the decompressed stream. Equation: output_bandwidth(codec) / decode_rates[codec]. Units: dimensionless core count; linear core scaling and identical cores are assumed.
output_bandwidth
engine.scenario_types.CompressedReadScenario.output_bandwidth(codec)Uncompressed byte rate generated by reading a named codec’s compressed data. Equation: read_bandwidth * compression_ratios[codec]. Units: bandwidth; the ratio is uncompressed/compressed size.
read_bandwidth
Type: Quantity.
Declared default: required.
compression_ratios
Type: Dict[str, float].
Declared default: required.
decode_rates
Type: Dict[str, Quantity].
Declared default: required.