srandom
| Input Stack: |
⇨ |
Output Stack: |
Generate a deterministic pseudo-random time series using a specified seed value. This creates
reproducible "random" data useful for experimentation, testing, and sample data generation.
Unlike :random, this operator allows creating multiple distinct but reproducible
random sequences.
Parameters
- seed: Integer seed value that determines the random sequence (different seeds produce different patterns)
Behavior
- Deterministic: Same seed always produces identical output for the same time range
- Hash-based: Uses hash of timestamp combined with seed for randomness
- Value range: Each datapoint is between 0.0 and 1.0
- Reproducible: Multiple queries with same seed and time range produce identical results
Examples
Creating a seeded random time series:
Seeded Random: /api/v1/graph?w=200&h=125&s=e-3h&e=2012-01-01T07:00&tz=UTC&q=42,:srandom
@@@
Seed Selection
- Different seeds: Produce visually distinct random patterns
- Consistent seeds: Ensure reproducibility across sessions
- Arbitrary values: Any integer works as a seed (positive, negative, or zero)
Comparison with :random
| Operator |
Reproducibility |
Multiple Sequences |
:random |
Single deterministic sequence |
No (always same pattern) |
:srandom |
Seed-based deterministic |
Yes (different seeds) |
- :random - Generate single deterministic pseudo-random sequence (no seed)
- :const - Generate constant values (opposite of random)
- :time - Generate time-based patterns
- :add / :mul - Transform random values to different ranges