On 04/04/2008, Jae Joo <jjoo@ecnext.com> wrote:
>
> Yes, I have same experience. The average time of AB is always faster than jmeter.
> AB is very light, but jmeter has it's overhead.
JMeter does have a higher overhead, but that is not the whole picture.
The overheads in JMeter are mainly to do with what happens between requests.
If I remember correctly, AB handles connections differently.
Also, is AB sending the same amount of data as JMeter?
> Specially, in client/server environment, it is different a lot.
>
> If you can, run AB and jmeter in the server - no network traffic overhead.
> The difference between jmeter and ab will be shrink down.
The network overhead will be much the same for AB and JMeter, unless
you are using JMeter in client-server mode.
>
> Jae
>
>
> -----Original Message-----
> From: Kishore_Kaushal@emc.com [mailto:Kishore_Kaushal@emc.com]
> Sent: Thu 4/3/2008 10:59 PM
> To: jmeter-user@jakarta.apache.org
> Subject: Difference in AB and JMeter Results.
>
> Hi,
>
>
>
> We are getting difference in response time results for same tests from
> JMeter and Apache-Benchmark.
>
> For same 10 users' response-time for single iterations:
>
> JMeter Response-time: 5000ms
>
> AB Response-time: 2500ms
>
>
>
> Please let me know if there is any possibility which is causing
> response-time delay from JMeter.
>
>
>
> PS: Both of the tests done from the same environment (client/server
> environment).
>
>
>
> Warm regards,
> Kaushal Kishore
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
|