Apache JMeter vs. AWS Device Farm

The key questions I would pose are: how high of a concurrent load do you need to generate against the system? How technically skilled are your personnel that will perform the load testing? How much programmable/scripting/code based control do you desire from the tool? Do you prefer more of a GUI based tool or a programmatic code/scripting based tool? JMeter will be for those that want a GUI, moderate to low scripting/code based control/programmability, moderate to low load generation, and for users experienced in the novice to expert range.

As long as you have the right scenarios in mind where you can use AWS Device Farm, you will be very happy. It's a cloud platform for testing automation support, where you can use real devices in multiple configs to validate your Android or iOS use cases. It does easily integrate with your CI pipeline, but reporting and UI are not perfect. A pain point is also the JUnit4 implementation, which could be more mature.

Incentivized

• Easy to construct Webservices script - You just need to add a sampler with the endpoint, payload and header information. And boom you can run a load test on the service with varying load and capture the response time and throughput. In a commercial tool like Loadrunner or Silkperformer, this process would take at least couple of hours, while with Jmeter we can do this in 10 minutes or less.
• Modularizing the scripts - Jmeter has a very good way of organizing the scripts and the reusability of the existing script (transactions) is also as easy as copy pasting the script.
• Jmeter tests can be run from anywhere, so you can run it from the idle webserver itself if you have some firewall restrictions to test it from your local or other environment.
• Jmeter distributed tests can be spawn from the AWS and results collated to see the performance from different regions in the world.

• Easy testing on many different devices
• Great integration with development workflow
• Customizable selection of testing framework

Incentivized

• Even though it has very easy installation, you can run into issues easily. I was ab;e to trigger a test at browser level but couldn't stop the tests. I had to kill it from task manager and then had to reinstall from begining.
• There are so many things in JMeter you can do especially with the sampler for selenium for UI tests. More documentation/webinars etc.. would help the community better.
• The jmx files should have an easy way to be integrated with ci/cd tools and cloud.

Incentivized

• Interface (CLI) not very easy. Competition has better options.
• Integration with IDE for test case coding.
• Reporting for the test scripts executed.

Incentivized

Price, Wiki and user sharing. Having access to the information provided by the developers and other open source providers is key for me. The ability to share information and get answers directly is very important to success in software testing. And the price of this product currently is amazing. Too many companies charge way too much money for products that are far behind in their value and pertinence

I can jump right into a new test plan and start building from scratch. The natural progression from test plan to thread group and then designing the basic format of the process is very streamlined and smooth. With only slight modifications I can build out a very complex model from a very basic beginning.

I have been using JMeter for the last year. By using this tool, you can make sure the system will work under varied loads. It helps us to simulate real time scenarios by creating required virtual users and make sure the application will work under load. Perform load, stress, and stability testing using JMeter.

Incentivized

We had the enterprise support with AWS, so overall support experience was good with great engineers on the back providing answers. As you may know, overall AWS support is different and this is not different. Responses through the regular web support channel came easily, fast and accurate. We had questions/issues which were solved fast. Documentation is good as well, especially around the test automation pieces.

Incentivized

I have evaluated LOADUI (web free version) and it was a very unstable tool and I could not rely on those results completely as I was not sure how the tool was performing. It only generated the top 10 less transaction times and when Jmeter was in use, it was very effective in using plugins to accommodate more outputs.

Incentivized

We haven't used anything like AWS Device Farm before. I am familiar with Amazon Web Services and when we had our MVP ready to test, we turned to AWS for a solution. AWS Device Farm was exactly what we were looking for as we have a really small team and limit resources.

Incentivized

• It helped to estimate resources required for a particular client requirement. For example if the client wants the application to be used by 500 users for 1 hour, based on the load test we perform for 500 users and 1 hour we recommend resources and their capabilities.
• Jmeter was used to identify bottleneck. This would lead to better performance.
• Jmeter was used to identify failures as well.

• It impact more and more. Not in terms of ROI but in time as well

Incentivized