What is our primary use case?
My use cases span various industries, from banking and energy to health care and now in telecommunications. The requirements encompass tasks such as load generation, finding response times, UI interaction, and identifying network issues.
How has it helped my organization?
LoadRunner Professional installation is pretty straightforward. I encountered challenges mainly when working for an energy organization as part of a COE (Center of Excellence).
In such scenarios, handling projects varied, sometimes involving SaaS platforms or terminal emulators for different protocols. Dealing with less common protocols could be tricky, and that's when we require support.
LoadRunner has improved our processes in the sense that once you record a script, even if you have to run a small generation, a developer can run it. They can even use it for their regression testing, you can use it. With parameterization, we can reuse LoadRunner to run in different environments just by changing the URL. You don't have to rescript.
LoadRunner is a tool you need to invest in beforehand. Once the installation is done, the major investment goes into scripting. You have to customize the script.
Once the script is customized, it's reusable. In a day, you might need to run multiple performance tests, so you can just keep running them as many times as you can through, like in the pre-prod environment, and then the prod environment. It definitely saves a lot of time.
Moreover, LoadRunner Professional has improved our product quality because, with the shift-left testing, we can detect early issues. We can run the test and then smaller bugs when it goes into production. The quality definitely improves.
It has also reduced the workload in the sense that when I was working in healthcare, we had to do a trend analysis report, and we wanted to run the test interactively. If we put the scenario in our auto-scheduler, we can check the report in the morning and we don't have to babysit it. We just run it.
What is most valuable?
In LoadRunner, the web protocol stands out as the strongest feature. It facilitates API testing and web load generation, and parameterization works exceptionally well.
We did not use TruClient for the load generation part. TruClient is a very heavy protocol, so it consumes a lot of memory and resources. Load generation cannot happen with TruClient protocols.
We would inject a few TruClient users with the load just to assess when the load is happening and what a TruClient is. Our use case was very different because we just wanted to know the user browser rendering time.
For us, it's not about how the users would use it for us, the use cases came from the UX designers. They would give us the requirement that if a page renders and you get a confirmation number, that is the last element to be loaded on the page.
The product offers good site performance under peak loads. We were launching these websites for the healthcare industry. Our response time from some of the pages was seven seconds, and these pages are very intense with all the medical equipment on the page. It came down to seven seconds, and then we were aiming to get to three seconds. So every time we would run the test, we would find issues. Sometimes, the page renders with a thousand rows of data, but you can paginate it so that the first page loads. The rest of it actually happens in the background, so it has drastically improved the performance of the pages.
There are not many products out in the market that support multi-protocol.
What needs improvement?
Everybody is going into using it on a non-Windows-based platform. Currently, it is Windows-based, so the product can improve that to support 64-bit and iOS Mac devices.
I recently just got to see LoadRunner Developer, but it is still not fully developed to use.
For how long have I used the solution?
I have been using this solution for around 17 years.
What do I think about the stability of the solution?
It is pretty stable. Issues arise only when a patch or something comes back.
What do I think about the scalability of the solution?
It's definitely scalable. It depends on the use case.
Which solution did I use previously and why did I switch?
When we started looking into getting LoadRunner licenses for AT&T, we picked three out in the market, like BlazeMeter and NeoLoad. We wanted to evaluate everything.
We gave a fair opportunity to all the other products so that every team member within our office could see everything.
But from what we saw, LoadRunner was cutting-edge. There were so many things with BlazeMater. We couldn't do what we needed to do with other products. With LoadRunner, the documentation sometimes is not updated but we get support when we are stuck. It's a great tool.
How was the initial setup?
The initial setup is very straightforward. You get the executables. You just need to get your hardware lined up, like which machine will be a controller and which machine will be a generator, whether you need two or three generators with whatever the requirement is.
You install the controller analysis on one machine, and then you have low generator agents, and different machines, and you just install the software, and the machine talks to each other.
What about the implementation team?
I installed LoadRunner Professional on the generators and the standalone machines.
What was our ROI?
We saw a return on investment. For example, if a production outage happens, we lose a lot of money. LoadRunner is a fraction of what we would lose.
It just pays off. We use the production downtimes as metrics for ROI.
What's my experience with pricing, setup cost, and licensing?
Shift-left is very new to our organization because we have been using professional/community licenses. We are trying to use enterprise now because, with professionals, it's a one-machine license. It's a single user.
We pay a lot for licensing. But the incentive is the support we get with it, that we pay once, and we are set.
What other advice do I have?
Overall, I would rate the solution a nine out of ten.