Skip to content

particula.execution.gpu_coupled_restart

Exact boundary restart for fixed GPU events and adaptive condensation.

Classes

NameDescription
GPUCoupledCheckpointSave event/RNG bytes and the complete deterministic boundary policy.

Functions

NameDescription
_validateCheck the complete host envelope without allocating device state.
checkpoint_coupledDownload exact continuation only after a successfully completed cycle.
load_coupled_checkpointLoad validated host records without physics or CUDA allocation.
resume_coupledRestore exact event streams and controls on the saved explicit device.
save_coupled_checkpointAtomically save canonical host bytes without pickle or uploads.

GPUCoupledCheckpointclassdataclass#

class GPUCoupledCheckpoint(schema_version: int, runtime: tuple, device: tuple, events: SimulationCheckpoint, configuration: object, controls: GPUCondensationControls, order: str, statistics: StepStatistics, counters: tuple[int, int, int, int], resident: bool)

Save event/RNG bytes and the complete deterministic boundary policy.

No in-progress adaptive controller is carried across a fixed event boundary: every cycle deliberately resets it from the new accepted physical state. Future forcing and source schedules are supplied again by the caller.

configurationattributeinstance attribute#

configuration: object

controlsattributeinstance attribute#

controls: GPUCondensationControls

countersattributeinstance attribute#

counters: tuple[int, int, int, int]

deviceattributeinstance attribute#

device: tuple

eventsattributeinstance attribute#

events: SimulationCheckpoint

orderattributeinstance attribute#

order: str

residentattributeinstance attribute#

resident: bool

runtimeattributeinstance attribute#

runtime: tuple

schema_versionattributeinstance attribute#

schema_version: int

statisticsattributeinstance attribute#

statistics: StepStatistics

__init__method#

def __init__(schema_version: int, runtime: tuple, device: tuple, events: SimulationCheckpoint, configuration: object, controls: GPUCondensationControls, order: str, statistics: StepStatistics, counters: tuple[int, int, int, int], resident: bool) -> None

_validatefunction#

def _validate(saved)

Check the complete host envelope without allocating device state.

checkpoint_coupledfunction#

def checkpoint_coupled(owner)

Download exact continuation only after a successfully completed cycle.

load_coupled_checkpointfunction#

def load_coupled_checkpoint(path, *, max_bytes=512 * 1024 * 1024)

Load validated host records without physics or CUDA allocation.

resume_coupledfunction#

def resume_coupled(saved, events, strategy, *, device)

Restore exact event streams and controls on the saved explicit device.

save_coupled_checkpointfunction#

def save_coupled_checkpoint(saved, path, *, compressed=True)

Atomically save canonical host bytes without pickle or uploads.