Time Stepping

Time Stepping

Goal

Provide a reliable workflow for choosing, ramping, and monitoring time steps plus output/checkpoint cadence.

Requirements

• Python 3.10+
• No external dependencies (uses stdlib)

Inputs to Gather

Decision Guidance

Time Step Selection

Is stability limit known?
├── YES → Use min(dt_target, dt_limit × safety)
└── NO → Start conservative, increase adaptively

Need ramping for startup?
├── YES → Start at dt_init, ramp to dt_target over N steps
└── NO → Use dt_target from start

Ramping Strategy

Script Outputs (JSON Fields)

output_schedule.py count is endpoint-inclusive: it includes both t_start and t_end, so count = number_of_intervals + 1 (e.g. t=0..5 at 0.05 spacing yields 101 frames for 100 intervals).

Workflow

1. Get stability limits - Use numerical-stability skill
2. Plan time stepping - Run scripts/timestep_planner.py
3. Schedule outputs - Run scripts/output_schedule.py
4. Plan checkpoints - Run scripts/checkpoint_planner.py
5. Monitor during run - Adjust dt if limits change

Conversational Workflow Example

User: I'm running a 10-hour phase-field simulation. How often should I checkpoint?

Agent workflow:

1. Plan checkpoints based on acceptable lost work:

   python3 scripts/checkpoint_planner.py --run-time 36000 --checkpoint-cost 120 --max-lost-time 1800 --json
   

2. Interpret: Checkpoint every 30 minutes, overhead ~6.7% (Acceptable per the interpretation table), max 30 min lost work on crash.

Pre-Run Checklist

• Confirm dt limits from stability analysis
• Define ramping strategy for transient startup
• Choose output interval consistent with physics time scales
• Plan checkpoints based on restart risk
• Re-evaluate dt after parameter changes

CLI Examples

# Plan time stepping with ramping
python3 scripts/timestep_planner.py --dt-target 1e-4 --dt-limit 2e-4 --safety 0.8 --ramp-steps 10 --json

# Schedule output times
python3 scripts/output_schedule.py --t-start 0 --t-end 10 --interval 0.1 --json

# Plan checkpoints for long run
python3 scripts/checkpoint_planner.py --run-time 36000 --checkpoint-cost 120 --max-lost-time 1800 --json

Error Handling

Interpretation Guidance

dt Behavior

Checkpoint Overhead

Verification checklist

• Recorded dt_recommended and dt_limit from timestep_planner.py and confirmed dt_recommended <= dt_limit with no "Recommended dt exceeds stability limit" note in the notes field.
• Captured the actual dt_limit value from the stability analysis (numerical-stability skill: CFL/Fourier/reaction limit) that was fed to --dt-limit, rather than guessing — and re-ran the planner after any parameter change.
• Confirmed safety <= 1.0 was applied (a margin below the limit), and logged the notes array (e.g. "Recommended dt reduced by stability limit", min/max clamps) so the binding constraint is known.
• Recorded the output_schedule.py count and verified it is endpoint-inclusive (count = intervals + 1, both t_start and t_end present), so frame counts and post-processing indices are not off-by-one.
• Recorded the checkpoint interval, method (daly vs cap), and overhead_fraction from checkpoint_planner.py, and confirmed overhead_fraction <= 0.10 (no warnings entry) against the overhead acceptability table.
• Confirmed every script exited 0 (not exit 2 / stderr ValueError) and that quoted dt/interval/checkpoint values come from the JSON results, not from a run that printed a validation error.

Common pitfalls & rationalizations

Security

Input Validation

• All numeric parameters (dt-target, dt-limit, safety, t-start, t-end, interval, run-time, checkpoint-cost, max-lost-time) are validated as finite positive numbers (non-finite values such as inf/nan are rejected)
• safety is bounded to <= 1.0 (a safety factor is a stability margin at or below the limit; values above 1.0 are rejected)
• ramp-steps and preview-steps are validated as non-negative integers with an upper bound of 1,000,000; only the previewed slice of the ramp is materialized to bound memory use
• Time range consistency is enforced (t-end must exceed t-start; checkpoint-cost must be less than run-time)

File Access

• Scripts read no external files; all inputs are provided via CLI arguments
• Scripts write only to stdout (JSON output); no files are created unless the agent explicitly uses the Write tool

Tool Restrictions

• Read: Used to inspect script source, references, and user configuration files
• Bash: Used to execute the three Python planning scripts (timestep_planner.py, output_schedule.py, checkpoint_planner.py) with explicit argument lists
• Write: Used to save generated time-step plans or checkpoint schedules; writes are scoped to the user's working directory
• Grep/Glob: Used to locate relevant files and search references

Safety Measures

• No eval(), exec(), or dynamic code generation
• All subprocess calls use explicit argument lists (no shell=True)
• Scripts use only Python standard library; no pickle loading or deserialization of untrusted data
• All output is deterministic JSON with no shell-interpretable content

Limitations

• Not adaptive control: Plans static schedules, not runtime adaptation
• Assumes constant physics: If parameters change, re-plan

References

• references/cfl_coupling.md - Combining multiple stability limits
• references/ramping_strategies.md - Startup policies
• references/output_checkpoint_guidelines.md - Cadence rules

Version History

• v1.2.2 (2026-06-24): Added Verification checklist and Common pitfalls & rationalizations sections grounded in the three planning scripts' actual outputs
• v1.2.0 (2026-06-23): Corrected overhead/frame-count docs and evals, removed output-time float drift, hardened input validation (checkpoint-cost < run-time, safety <= 1.0, bounded ramp/preview steps, finite checks)
• v1.1.0 (2024-12-24): Enhanced documentation, decision guidance, examples
• v1.0.0: Initial release with 3 planning scripts