How to get the TPS value

Hi,
I am new to Gatling. I was using Jmeter before. I am able to get the TPS value in JMeter reports but the same is missing in the latest versions of Gatling.

I could see that the Transactions per second over time graph was present in Gatling 2.0.0-RC2 though http://gatling.io/docs/2.0.0-RC2/general/reports.html .

How to get this info in latest version?

regards
Krishna

Reports in Gatling 2.2.2 have two tabs (Global and Details).

Global tab shows overall simulation stats whereas “Details” tab shows graphs/stats for each transaction/group.

You can find the requests/sec in the “Details” tab for each transaction in your simulation.

http://gatling.io/docs/2.2.2/general/reports.html?highlight=reports#requests-per-second-over-time

Hi,
Thanks very much for your answer. I have two questions though.1. Is it comparable to the TPS value we get in Jmeter reports?
2. Should we consider the Requests per second or Response per second charts?

Thanks!

regards
Krishna

  1. I am not sure which Jmeter reports you are talking about coz there are lots of plugins available. I can see Gatling gathers very limited amount of data, how many requests are made per any given time of the execution, are the responses OK or KO, what time each request and response take. All this information is stored into simulation.log file. Whereas Jmeter has logs of plugins to get more detailed information. Having said that, the stats reported by both tools should be identical as long as the test setup is similar.

  2. Requests per second graph shows how the load is being pushed on the server. Response per second indicates how the server responded to those requests i.e. total, success and failures
    If there are zero failures in the test, both graphs should look identical. I hope I put it correctly :slight_smile:

Note that the requests per second graph corresponds to the ACTUAL number requests triggered by the script, it is not necessarily the number of transactions that you would have defined in the script. I know this statement is a bit confusing. So example comes to rescue.

`
startTransaction
httpRequest1
httpRequest2
httpRequest3
EndTransaction

`

One execution of above will show 3 requests in request per second graph, unlike tTPS graph in Loadrunner which will show only 1 request (for your transaction). This is what I observed.
@Stephane, thoughts?

What you call transactions is LR is what we call groups.
One thing can cause the number of actual requests to not match what you could expect: caching.

Thanks all for your comments.
@CyberNinja: I was referring to the Throughput value in http://jmeter.apache.org/usermanual/component_reference.html#Aggregate_Report

regards
Krishna