Can i use k6 output to measure RPS as output in terminal?
Hi @Elibarick
Answering the question from the topic title:
Difference between Request Per Second vs Iteration per second?
So the difference is that iteration can contain many requests.
Can i use k6 output to measure RPS as output in terminal?
Could you please elaborate on your question? I’m not sure if I’m following 
Thanks!
1 Like
@olegbespalov
This is not to hijack this thread but I am also curious about this or something related to it;
ramping-vus vs. ramping-arrival-rate
This is my understanding:
Ramping-vus: Each VU will try to execute as many iterations as possible over a given amount of time.
Ramping-arrival-rate: Will try to execute given amount of iterations at a given rate using the amount of VUs allocated.
What I don’t understand is that with ramping-vus you can reach a high request per second with very low iterations per second, but you cant do that with ramping-arrival-rate - or what?
To elaborate, I tried to create the exact same test that I had using ramping-vus executor but with ramping-arrival-rate instead but I just cannot get the same results.
Hi @agan
What I don’t understand is that with ramping-vus you can reach a high request per second with very low iterations per second, but you cant do that with ramping-arrival-rate - or what?
It’s probably because your VU executes more than one request in a single iteration. And for the ramping-arrival-rate where the number of the iterations grows differently from the ramping-vus RPS is growing slowly.
Hello @olegbespalov
k6 output result show value RPS and Iteration/sec. I want to known result of each requests RPS and Iteration instead of all scenario result total.
I want to known result of each requests RPS and Iteration instead of all scenario result total.
The most proper way of doing this is using some metrics storage and outputting k6 there for the later investigation.
For the default k6’s output, it’s not possible without workarounds. There are a bunch of similar topics we have on this forum that basically answers how to implement them:
- Separate metrics summary for each request in default function
- How to track request duration for each endpoint?
- Graphql test serveral stages and every stage have multiple request - #3 by olegbespalov
Let me know if that answers,
Cheers!