WAPT vs LoadRunner: Which suits you better?
HP LoadRunner is one of the pioneering tools in the load testing industry. While most of its counterparts had left the market long ago to free space for modern solutions, LoadRunner still remains very popular among testing professionals. Does this necessarily mean that new users should follow the tradition? How the choice between LoadRunner and WAPT Pro is seen from the new user perspective?
 |
LoadRunner is probably the most expensive solution on the market. If you compare WAPT with any other load testing tool, you will probably find that for the majority of applications WAPT-based solution is more cost-efficient. However with certain test requirements and environment there is a chance for another tool to be somewhat cheaper but only if that tool is not LoadRunner. There is no way to save with it compared to any competitor. Each and every test you run with LoadRunner will cost you more than the same test conducted with WAPT or even other tools. The license price would create that difference by itself, but surprisingly there are extra reasons for that. If you think that after paying a lot for the license you will get an elaborated solution that will let you save resources during the tool usage, you are wrong. In fact, LoadRunner imposes very high requirements on your QA team and processes, which means that you need to hire more qualified engineers and plan more time for testing. |
| LoadRunner offers the widest functionality among all load testing tools. It is the absolute winner in this respect. Does this create a real benefit? Yes, if you are constantly working on the most complex and sophisticated tests in the universe. LoadRunner claims to support tens different protocols. Half of them are HTTP-based, which means that they are actually supported by any other testing tool that works over HTTP. Second half comprises protocols that are so rarely used that you will hardly need to run a test with a single one of them in your entire career. This means that for the majority of tests this extra functionality is not required at all. Does this mean that you can easily put it away and use LoadRunner the same way you would use WAPT? No. Imagine yourself in the cockpit of a fighter jet sitting before the instrumental panel. You may dream to get there and try a flight one day. This may be quite a fascinating pastime. Now imagine that you have to use a fighter jet instead of your car to drive to a supermarket or wherever you need during a day. That is exactly what your regular use of LoadRunner will look like compared to the use of a less impressive, but neat piece of software, such as WAPT. |  |
 |
Here we come to the discussion about usability. You may be surprised to know that LoadRunner is a scripting tool. This means that every test is a script. It can be initially created by the tool after you record a user session with a browser. However, all your further work on the test is done through making modifications to that huge script file. In other words, you need to be a programmer to be accustomed to this process. Unfortunately, even for people with good programming background such approach does not serve making things neither easier nor faster. This means longer time required for the support of the regression tests and for the design of new ones. This also creates a problem with the verification of test results. While LoadRunner is widely recognized as a tool creating an accurate emulation, it is not always easy to understand what a particular test emulates unless you are a professional tester. This basically makes it a tool designed for developers not only in terms of interface, but in terms of delivery as well. This resembles writing load tests in C++ or Java, which is a usual starting point for a development team just becoming aware about the performance problems in general. |
| Load generation may require significant hardware resources. That is why it is important to utilize them efficiently, especially in case you need to generate a really high number of concurrent virtual users. This is another reason why too sophisticated functionality and too general approach to the load generation may become a drawback. With average test options and hardware used to run the load generators, you may expect to create about 1,000 concurrent virtual users per system with LoadRunner. A single WAPT Pro load engine being highly optimized for the specific web testing tasks can run 10,000 or more in the same situation. This is a 10 times difference in performance, which may become crucial, if you consider hardware costs. |  |
 |
Evaluating this from another perspective you can easily see that running really high load test with LoadRunner is a troublesome task. With WAPT there is a little difference between generating 100 users and 10,000 users. You will only need to replace regular agent with a more productive engine component. If you need more, you can easily add more engines or get a better hardware to increase the capacity of each engine. Distributed testing is implemented in both tools, and it works smooth enough if you add several nodes. But how many ones you will need to add in order to run 100,000 users with LoadRunner? Isn't it safer to stick to configurations with countable parts to have some visibility? |
| It is quite reassuring to feel that a huge international company, like HP, is backing your testing efforts. The problem is that they have thousands clients like you and would not care to lose one. This does not mean that they treat their clients badly. They just cannot recognize your task as a very special one. They will provide a bulk support service to you, and that may work quite well while you are stumbling at common obstacles well known to them. However they simply cannot pay too much attention to your requirements, if you need to fix or improve anything in the product or if you need somebody to look into your specific test implementation, unless you are ready to pay enormous fees for such additional services. This is all completely different with WAPT, which is supported by a small, but very experienced team of professionals who really care about every client, existing or prospect. This is not a mere form of attitude, but a very practical thing: if you happen to find a functional problem, you can expect a personal fix in a day or two; if you need a new feature to facilitate your work, you can expect it to be thoroughly considered and implemented in terms of days or weeks. |  |
Conclusion
In summary, LoadRunner is a great tool for you, if all of the following are true:
• You have unlimited budget to buy licenses, hardware and to pay your QA team for extra time they spend designing and supporting tests.
• Your company has well-established processes in the development cycle ensuring all activities are made strictly according to the plan providing sufficient time for each step.
• You and everyone else involved in the test design and execution has a good programming background.
LoadRunner customers seeking for a more affordable solution with sufficient functionality to replace their current "hi-end" burden are quite common newcomers among WAPT clients. On the other hand, we have never heard from anyone about a switch from WAPT to LoadRunner. Even if you need to test an exclusively rare protocol one time a year, it will be more rational to use an on-demand LoadRunner cloud generator for that purpose, while running your regular web tests with a more affordable, fast and easy to use solution, such as WAPT.
Get WAPT demo
Get tool comparison for your project
