Search Ctrl K
Home / Tools
THE FEUTEX TOOLKIT

Training Checkpoint IO Budget

Estimate checkpoint pause overhead and storage occupied by retained checkpoints.

AI Infrastructure workspace● Runs in your browser
01 Your input
02 Your result
View or copy raw result data
FormulaSynchronous checkpoint pause = size/speed; Wall time spent checkpointing = 100*(size/speed)/(interval*60+size/speed); Retained checkpoint storage = size*keep

Made for real tasks

  • Estimate checkpoint pause overhead and storage occupied by retained checkpoints.
Explore step-by-step workflows →

Worked example

Inputs

Checkpoint size
40 GB
Write throughput
2 GB/s
Compute between checkpoints
30 min
Checkpoints retained
5 checkpoints

Expected results

Synchronous checkpoint pause
20 s
Wall time spent checkpointing
1.098901099 %
Retained checkpoint storage
200 GB

Load the example in the workspace to reproduce these results. Display rounding does not change the underlying calculation.

How to use this tool

  1. Enter checkpoint size, write throughput, compute between checkpoints, checkpoints retained in the displayed units.
  2. Calculate the scenario, then compare it with measurements before changing a system.

Try this example

{
  "size": 40,
  "speed": 2,
  "interval": 30,
  "keep": 5
}

Accuracy and scope

Synchronous nonoverlapping writes; asynchronous checkpoints may hide some overhead.

Is my data uploaded?

This tool runs locally in your browser. Input and output are not sent to a processing server. Copy or download your result before leaving the page.