0 ms Minimum Response Time?

[lo...@fastmail.fm]

Hi, 

I am using an Aggregate Report listener. I have a minimum HTTP response
of 0 ms. But how can this be so? How can a web transaction be less than
1 ms?

Many Thanks

Paul


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org

Re: 0 ms Minimum Response Time?

[sebb <se...@gmail.com>]

On 1 May 2014 16:24, John Schulz <jo...@aol.com> wrote:
> Any chance we can make a change to JMeter to support reporting in
> microseconds? I have run into a few projects in the last couple of years
> where I would have used JMeter if only it could report sub-millisecond
> response times.  I have done a few tests using the java sampler to see how
> many calls I can make to a sampler in one millisecond and on my laptop I was
> getting about 50 calls per millisecond. So microsecond response time might
> be over kill but at least I could measure those 50 calls without having to
> do some averaging.

JMeter already supports using System.nanoTime().
However, this has some drawbacks as it tends to drift quite
significantly from System.currentTimeMillis().
It's also more expensive.

The sampleResult class currently stores elapsed time in milliseconds.
I don't know if it would be feasible to change/enhance that without
breaking existing tests.

I suggest you raise an enhancement request.

> -----Original Message-----
> From: loadtest@fastmail.fm [mailto:loadtest@fastmail.fm]
> Sent: Thursday, May 01, 2014 10:53 AM
> To: user@jmeter.apache.org
> Subject: 0 ms Minimum Response Time?
>
> Hi,
>
> I am using an Aggregate Report listener. I have a minimum HTTP response of 0
> ms. But how can this be so? How can a web transaction be less than
> 1 ms?
>
> Many Thanks
>
> Paul
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
>

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org

RE: 0 ms Minimum Response Time?

[John Schulz <jo...@aol.com>]

Any chance we can make a change to JMeter to support reporting in
microseconds? I have run into a few projects in the last couple of years
where I would have used JMeter if only it could report sub-millisecond
response times.  I have done a few tests using the java sampler to see how
many calls I can make to a sampler in one millisecond and on my laptop I was
getting about 50 calls per millisecond. So microsecond response time might
be over kill but at least I could measure those 50 calls without having to
do some averaging.

-----Original Message-----
From: loadtest@fastmail.fm [mailto:loadtest@fastmail.fm] 
Sent: Thursday, May 01, 2014 10:53 AM
To: user@jmeter.apache.org
Subject: 0 ms Minimum Response Time?

Hi, 

I am using an Aggregate Report listener. I have a minimum HTTP response of 0
ms. But how can this be so? How can a web transaction be less than
1 ms?

Many Thanks

Paul


---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org

Re: 0 ms Minimum Response Time?

[sebb <se...@gmail.com>]

On 1 May 2014 16:19,  <lo...@fastmail.fm> wrote:
> Hi Sebb,
>
> It's probably because I am running tests within the same network, I
> shouldn't hit caching as I am using different variables per request. I'd
> expect at least 1 ms.
>
> No, I am not using HTTP Cache Manager?
>
> Can the timer resolution be made more granular?

Timer resolution is generally a function of the OS, also the hardware
may be relevant.

> Many Thanks
>
> Paul
>
> On Thu, 1 May 2014, at 08:00 AM, sebb wrote:
>> On 1 May 2014 15:53,  <lo...@fastmail.fm> wrote:
>> > Hi,
>> >
>> > I am using an Aggregate Report listener. I have a minimum HTTP response
>> > of 0 ms. But how can this be so? How can a web transaction be less than
>> > 1 ms?
>>
>> Possible causes:
>> - Fast webserver/network
>> - The timer resolution may be insufficient to measure the response
>> time accurately.
>> - Are you using HTTP Cache Manager?
>>
>> > Many Thanks
>> >
>> > Paul
>> >
>> >
>> > ---------------------------------------------------------------------
>> > To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
>> > For additional commands, e-mail: user-help@jmeter.apache.org
>> >
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
>> For additional commands, e-mail: user-help@jmeter.apache.org
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
>

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org

Re: 0 ms Minimum Response Time?

[lo...@fastmail.fm]

Hi Sebb, 

It's probably because I am running tests within the same network, I
shouldn't hit caching as I am using different variables per request. I'd
expect at least 1 ms. 

No, I am not using HTTP Cache Manager?

Can the timer resolution be made more granular? 

Many Thanks

Paul

On Thu, 1 May 2014, at 08:00 AM, sebb wrote:
> On 1 May 2014 15:53,  <lo...@fastmail.fm> wrote:
> > Hi,
> >
> > I am using an Aggregate Report listener. I have a minimum HTTP response
> > of 0 ms. But how can this be so? How can a web transaction be less than
> > 1 ms?
> 
> Possible causes:
> - Fast webserver/network
> - The timer resolution may be insufficient to measure the response
> time accurately.
> - Are you using HTTP Cache Manager?
> 
> > Many Thanks
> >
> > Paul
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> > For additional commands, e-mail: user-help@jmeter.apache.org
> >
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org

Re: 0 ms Minimum Response Time?

[sebb <se...@gmail.com>]

On 1 May 2014 15:53,  <lo...@fastmail.fm> wrote:
> Hi,
>
> I am using an Aggregate Report listener. I have a minimum HTTP response
> of 0 ms. But how can this be so? How can a web transaction be less than
> 1 ms?

Possible causes:
- Fast webserver/network
- The timer resolution may be insufficient to measure the response
time accurately.
- Are you using HTTP Cache Manager?

> Many Thanks
>
> Paul
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> For additional commands, e-mail: user-help@jmeter.apache.org
>

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
For additional commands, e-mail: user-help@jmeter.apache.org