Distributive Testing

What is Distributive testing and why is it required?

When a single machine is unable to simulate enough users to stress, more than one machine is used to perform the test (two or more computers will interact with each other). Using multiple machines to perform the test is called as Distributive testing. The crucial points of distributed tests are synchronization, administration, and configuration.

Configuration of Distributive testing in JMeter

1. Install the JMeter in all the machines that are used for testing. Make sure to use the same version of JMeter on all the machines as mixed versions may not work correctly.

2. Client and server machines should be on same subnet.

3. JMeter client machine should able to access the server machines

The JMeter master controller machine will initiates the test on multiple server machines. The Master machine will act as a console.

Steps

1.       In Master machine, open jmeter.properties file from JMeter/bin directory

2.       In jmeter.properties file, search for remote_hosts (the default value will be 127.0.0.1)

3.       Add the IP address of the server machines with comma(,) as the separator

4.       In our case the entry should be as below

remote_hosts=192.168.50.2,192.168.50.3,192.168.50.4,192.168.50.5

Below is the screen shot from jmeter.properties file

5.       On all server machines, execute jmeter-server.bat from JMeter/bin directory.

The dos window will appear with “jre\ [version]\bin\rmiregistry.exe”.

6.       Come back to Master machine and open the JMeter and open the test script that need to test

Click on RunàRemote start

All the server IP address can be seen in Remote start option as shown below.

To run the test on single server, Click on Run -->Remote start -->ip address of the server (ex: 192.168.50.2)

To run the test on all the servers, Click on Run-->Remote start All

Note: All the results will be sent to Master machine from the servers and so the network IO on Master machine will get saturated. Its better to use simple data writer to save the results and view the file after the test execution is completed.