On 29/10/2009, Ronan Klyne <[email protected]> wrote: > Carl Shaulis wrote: > > > Hello, > > > > By the way this is an active and extremely responsive forum, so THANK YOU! > > > > I have a question regarding some data observations. We ran a load test > > using a script with 5 thread groups. We save the data into a .csv file > then > > import the data into MySQL. > > > > Some of the requests show 40K bytes, but have a latency that equals 0. > > > > Discussing this amongst the team we conclude that these requests are being > > redirected. > > > > I guess I can understand why the latency might be zero, but I am not sure > > why the report indicates a large file size. > > > > > > > I thought the latency was the time to the first byte, and the elapsed time > was the total. See > http://jakarta.apache.org/jmeter/usermanual/listeners.html#csvlogformat > > If I'm wrong, I'd like to be corrected...
You are not wrong. Strictly speaking, latency is time to first response, which will in general be more than 1 byte. JMeter relies on Java and the OS to do the buffering, so can't easily check time to first byte. Note that some samplers don't measure the latency, in which case it will always be zero. > Ronan > > > Any thoughts? > > > > Since I have your attention does anyone have a baked SQL script for > > returning average latency per thread group? :o) > > > > Thanks, > > > > Carl > > > > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: > [email protected] > For additional commands, e-mail: > [email protected] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
