Skip to main content

Send a fixed number of requests per second

Real traffic does not slow down when your server does. The constant-arrival-rate executor keeps the rate fixed, so a slow server shows up as a growing delay, not as fewer requests.

vegaload run -target http://127.0.0.1:8080/widgets -protocol http1 \
  -executor constant-arrival-rate -rate 50 -max-vus 100 -duration 30s

-rate is iterations per second. -max-vus is the most users VegaLoad may use to keep that rate.

Make the rate climb, spike and fall​

A flat rate shows whether the target holds one level. To see how it behaves as the load changes, use ramping-arrival-rate. -stages are rate:duration pairs. The rate moves in a straight line from where it was to each target, starting from -rate (default 0):

vegaload run \
  -executor ramping-arrival-rate -max-vus 200 \
  -stages 50:4s,200:4s,600:3s,200:4s,50:3s \
  scenario.vl.js

Like constant-arrival-rate, it keeps starting iterations on schedule even when the target slows down, and drops an arrival when -max-vus are all busy. -cap-rate applies to the highest rate in the stages.