Pages

Monday, 30 January 2017

JMeter - Timer - "Gaussian Random Timer"

As the name suggested, Gaussian Random Timer is used to generate the random delay. This timer is based on Normal or Gaussian Distribution Function. The delay (think) time is the sum of the Gaussian distributed value (with mean 0.0 and standard deviation 1.0 ) times the deviation value you specify and the offset value. 

How to add "Gaussian Random Timer" element?
You can follow below steps:
1. Select "Test Plan" or "Thread Group" or "Controller" or "Sampler" node where you want to add the timer
2. Right-click on the node
3. Mouse hovers "Add"
4. Mouse hovers "Timers"
5. Click "Gaussian Random Timer"

How to remove "Gaussian Random Timer" element?
You can follow below steps:
1. Select "Gaussian Random Timer" node
2. Right click on "Gaussian Random Timer" node
3. Click "Remove" (Mouse) or "Delete" button from the keyboard
4. Click "Yes"    

What are the attributes of "Gaussian Random Timer" element?

"Gaussian Random Timer" has following attributes:
  • Name: To provide the name of the timer
  • Comments: To provide arbitrary comments (if any)
  • Deviation (in milliseconds): The number shows how much the delay can deviate from the given offset towards higher and lower range. 
  • Constant Delay Offset (in milliseconds): The number shows a constant delay which will be added in random number generated by Gaussian Function in the range of given deviation value.
Example: If Constant Delay Offset is given 1000 and deviation is given 200 then Gaussian Random Timer will generate the random number in the range of 800 to 1200 milliseconds:

Gaussian Random Timer Delay = 1000 + GF*200   (where GF value could be from -1 to 1)
The probability of satisfying the deviation range is around 68%. 

Next Topic:
Apache JMeter - "JSR223 Timer" Element

Previous Topic:
Apache JMeter - "Constant Timer" Element

Related Topics:

3 comments:

  1. Hi,
    I did not understand below points-
    1. Is 68% deviation range is predefined or we have calculated it?
    2. Also if i have loop count as 5 ,so for all 5 iteration delay will be in between 200-400?

    ReplyDelete
  2. Hi Pallawi,

    1. 68% is the predefined. If you need to get more information then please refer Gaussian Distribution Theory.

    2. I am not sure what the value of offset you have considered here.

    As given in the example if offset is 1000 and deviation is 200 then you will get values in the range of 800 to 1200. All the 5 loops may have different values in this range.

    ReplyDelete