Files
cs249r_book/mlsysim/docs/api/core.solver.CheckpointSolver.qmd
Vijay Janapa Reddi 2bbe3e1a69 docs(mlsysim): redesign website, add 12 tutorials, and CLI entry points
Replace 9 old tutorials with 12 new numbered tutorials (00-11) covering
roofline through full-stack audit. Redesign landing page, add
models-and-solvers and extending-the-engine guides. Add __main__.py,
cli.py, and cli/ package for command-line interface.
2026-03-12 16:04:51 -04:00

50 lines
1.6 KiB
Plaintext

# core.solver.CheckpointModel { #mlsysim.core.solver.CheckpointModel }
```python
core.solver.CheckpointModel()
```
Analyzes the storage constraints and I/O burst penalties of saving model states.
Training massive models requires saving hundreds of gigabytes (weights +
optimizer states) to persistent storage. This solver calculates the time
spent blocked on I/O, subtracting from the cluster's Model FLOPs Utilization.
Literature Source:
1. Eisenman et al. (2022), "Check-N-Run: A Checkpointing System for
Training Large Language Models."
## Methods
| Name | Description |
| --- | --- |
| [solve](#mlsysim.core.solver.CheckpointModel.solve) | Solves for checkpoint size, write time, and resulting MFU penalty. |
### solve { #mlsysim.core.solver.CheckpointModel.solve }
```python
core.solver.CheckpointModel.solve(
model,
hardware,
optimizer='adam',
checkpoint_interval_hours=4.0,
)
```
Solves for checkpoint size, write time, and resulting MFU penalty.
#### Parameters {.doc-section .doc-section-parameters}
| Name | Type | Description | Default |
|------|------|-------------|---------|
| model | Workload | The model architecture. | _required_ |
| hardware | HardwareNode | The target hardware for storage bandwidth. | _required_ |
| optimizer | str | Optimizer type ('adam' or 'sgd'), determines bytes per parameter. | `'adam'` |
| checkpoint_interval_hours | float | Hours between checkpoints. | `4.0` |
#### Returns {.doc-section .doc-section-returns}
| Name | Type | Description |
|------|------|-------------|
| | CheckpointResult | Checkpoint size (GB), write time, storage bottleneck flag, and MFU penalty percentage. |