https://github.com/bluekeyes/rndout

Generate random output for testing throughput
https://github.com/bluekeyes/rndout

Last synced: 3 months ago
JSON representation

Generate random output for testing throughput

• Host: GitHub
• URL: https://github.com/bluekeyes/rndout
• Owner: bluekeyes
• License: mit
• Created: 2022-06-10T22:07:13.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2024-09-20T22:06:59.000Z (9 months ago)
• Last Synced: 2025-01-13T21:33:03.863Z (5 months ago)
• Language: Go
• Homepage:
• Size: 5.86 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# rndout

Generate random output for testing. Originally written to simulate output from

builds while testing a CI system. May be useful for other things that just need

throughput and not a specific data format.

## Usage

```

$ rndout -help

  -block-size int

        maximum number of characters printed in one line/operation (default 4096)

  -duration duration

        duration (default 1m0s)

  -mode string

        the operation mode, one of 'logistic' or 'ramp' (default "logistic")

  -ramp-duration duration

        time taken to reach the peak rate; only used with -mode=ramp (default 10s)

  -rate string

        peak character rate in chars/s (default "128")

  -scale int

        scale factor for the output distribution; only used with -mode=logistic (default 25)

  -skip-probability float

        probability that a given slice will contain skips

  -skips int

        expected number of time steps with no output per slice (default 2)

  -slice-length int

        number of time steps per slice (default 16)

  -step-size duration

        length of each time step (default 250ms)

```

Output is written to `stdout`.

## Algorithm

### `ramp` mode

Linearly increase the output rate on each step until reaching the peak output

rate after the ramp duration. Remain at that output rate for the remaining

time.

### `logistic` mode

1. Divide the duration by the step size

2. Select a random step at which to reach the peak output rate

3. For each step, print random ASCII characters such that the output rate

   follows a logistic distribution with scale `scale` centered at the peak step

4. Every `slice-length` steps, randomly sample a Poisson distribution to

   determine how many steps to skip printing output. This reduces the actual

   output rate but can add more realistic pauses and gaps in the output.

## License

MIT