WAPT vs JMeter: Which suits you better?


Apache JMeter is one of the most widely known load testing tools, not least because it is free and open-source. This sounds attracting, but unless zero license cost is your only target, most likely you will be better off with WAPT Pro. Here are 7 key reasons why.

Cost of Possession
The cost of possession is not just the amount of money you spend when buying a license. If you take into account time spent by each of your team members getting familiar with the tool, configuring it, creating tests, supporting them and analyzing results, you will see that JMeter is actually an expensive and doubtful solution in comparison with many commercial tools on the market.
  
Excessive amount of information about JMeter does not guarantee a steep learning curve, as you will have to spend even more time selecting the right sources. Strange and confusing terms used in the tool to represent very common notions (like "thread" actually denoting a virtual user) do not serve better clarity. WAPT users spend much less time getting familiar with the tool, because we provide relevant and efficient guides as well as personal support and assistance throughout the whole cycle of product evaluation and use. Finally, WAPT user interface is much more intuitive, so you do not need to get through hundreds menu items and remember the right ones to create a simple test.
Learning Curve
  
Test Design
JMeter generally inherits the concept of old-fashioned scripting tools that required you to program each test from scratch. The step by step representation of the user session in form of HTTP requests is mixed with elements of different types with strange names, like "listeners". Overall this still looks like a structured script. The test recording is an option, not quite an obvious one. Most JMeter users prefer third party recording tools, which is not surprising. When it comes to parameterization and correlation, you can do this, but with no automation and you will need to know really well what you are doing, because even to extract a single value between two boundaries, you will need to use regular expressions.

WAPT applies straightforward test design concept: first, you record a user session; second, you adjust all session-specific values in it, so that it would be possible to replay its copies in the test. Majority of work is done automatically. A number of features are provided to let you do all the rest with ease: automatic search and parameterization, side by side comparison using highlights, extraction test on the initially recorded content, etc. This functionality lets you create complex tests 2 to 5 times faster with WAPT then with JMeter. If you need to rebuild your tests after a website update, you may do this even 10 times faster, because you can set automatic rules applied on recording.
  
All JMeter features are accessible through its huge menu consisting of several levels. Features that you need every day are not anyhow easier to access than features that you almost never need. Default feature set is not sufficient to create an average test, so you have to look for plugins. There are tens, if not hundreds, of them. Each one is installed separately and adds something to that huge menu. This becomes even harder to configure when you need to run a distributed test with multiple systems, because each of them should have exactly the same JMeter configuration installed.

WAPT is a solution designed to guide you through the test design and execution cycle with as little overhead as possible. On each step you have primary access to the features you need at the moment. Advanced features are available as replacements for the basic ones, like JavaScript processing instead of simple boundaries based extractor.

Installation of WAPT Pro with full module pack takes less than 3 minutes. You can start using the tool immediately after that. Should you need to attach a cloud instance for running a higher load, all you have to do is to launch it at AWS and specify its address in WAPT Pro.
Tool Configuration
  
Capacity and Performance
Our tests have shown that JMeter and WAPT can generate approximately the same number of page requests per second on a static website using the same hardware. This is not too bad for JMeter, taking into account that it is written in Java. However JMeter underperforms WAPT significantly on complex tests with a big number of session-specific values (which actually means most real tests). This becomes even worse, if you try applying all the features you see in JMeter tutorials, because many of them require a lot of memory and CPU resources.

WAPT performance also depends on the options you set for the test, but all those options are practically applicable for every test and do not produce surprises. For example, if you enable full logging, you should have HDD interface working at the same rate as your network, but that is an obvious requirement.
  
To get any test results in JMeter you need to select what exactly you would like to see before running the test. You have a choice of multiple menu items again. If you add all possible options, you risk overloading your own system with calculations. Yet, none of them may be particularly useful for your test. That is why JMeter “professionals” recommend dumping everything to a CSV file and use third party tools to create a real report. There are some plugins that can create files with charts. Surely, in the end you can get everything required to create a good test report, but you will spend hours looking for right plugins, configuring them and compiling everything together.

In WAPT you get a single comprehensive test report. It is structured and customizable, but you do not need to spend time configuring the tool just to generate it. The report includes all the relevant test data. Performance counters for each of the configured servers as well as pass/fail criteria results are also included. Interactive charts with adjustable sizes can be accessed by selecting the corresponding left view items. More detailed information can be found in logs shown in the integrated viewer available through the same view. Those logs comprise not just tons of text messages, but fully structured items with easy navigation.
Test Results
  
Support and Services
The choice of JMeter basically means that you are ready to face any problems and deal with them without any external assistance. Our approach is different, as we believe that fast and efficient support service is a key component not only for novice users, but for experienced professionals as well. This includes answering questions on the tool functionality as well as specific solutions depending on the situation. This really saves time on all stages of the process. Should you face a functional problem in WAPT, you can expect us to fix it within days, if not hours. In case of JMeter you will hardly have any chance of getting any fixes at all. Finally, we do guarantee the quality of our tools and the accuracy of test results obtained with help of them.

Conclusion

JMeter is a great tool for you, if all of the following are true:

  • You need to test a site with few pages and you have unlimited time for that task.
  • You just love to spend time configuring software and getting everything out of it step by step.
  • You always do everything on your own and hate asking for assistance.
  • You are ready to fix Java code in case you face a problem in the tool.
  • You do not need any guarantee from anyone that your tool works correctly, because you fully understand the technical background behind each feature.

If you have doubts regarding any of the above points, you are welcome to choose WAPT. It will let you design and update your tests faster and make the whole process more predictable, as it will be backed by our own QA team. You can do it completely free using the demo version.