OpenText LoadRunner Professional Reviews

All of our installations are on-premises at the moment.  

One of the most valuable features to me is that LoadRunner is familiar. I have worked with it for so long that I know where everything is and I know how it is supposed to work.  

But focusing on actual program features, all I can say is it has good protocol coverage.  

I have not had a really good look at the newest versions of LoadRunner. The problem I have always had with LoadRunner — and even more so with Performance Center — is that it is not very good with agile delivery and it is very difficult to integrate. Software engineers who have been working in agile delivery have been saying this for some time, and have been having success with alternatives.

Also, Performance Center has historically been quite unreliable and difficult. It tends to fail at collation. I think that is because of the Load Runner architecture that Performance Center inherited. Everything waits until the end of a test to collate and it does not always work out well. I know that Micro Focus has done something about that in the new version. But that is the worst thing that happens: with LoadRunner or Performance Center you run a big critical test and you can't get the results.  

I have personally been using LoadRunner and Performance Center for 20 years. I think it's been the backbone of performance testing for 20+ years and must be given credit for helping establish the discipline.

Stability has been an issue. It has really been a big problem for us with Performance Center in particular. Crashing or exiting during collation is not acceptable.   

We have not been totally satisfied with the scalability. We tend to have more issues with Performance Center. The more you try to grow use with it on-prem, the more headaches you have. Often working with LoadRunner and smaller environments is more reliable. We believe the SaaS solution might be better because you have got centralized management.  

Most of my experience is in the early days of Mercury, they were brilliant. They were really good and helped resolve issues. I have not been using them recently.  

Right now we use LoadRunner and Performance Center depending on what we believe are best practices for different situations. We are considering evaluating other options as well, including LoadRunner SaaS.

The two Micro Focus products are interesting to compare. LoadRunner is very easy to set up. Performance Center needs an expert. I can install LoadRunner in minutes by myself. To do the installation properly with Performance Center needs a specialist and takes weeks.  

Plenty of people seem to have installed Performance Center in a couple of days, but when they use it they usually have problems. You really need to do it properly. I do plan to have a look at the newer versions, and maybe those implementations will be smoother.  

Advice that I would give people considering LoadRunner is that I would recommend exploring other tools first or at least in comparison. There are lots of really good open-source or even just cheaper alternatives. Depending on your use case, the other options might be much better. LoadRunner has broad protocol coverage. Sometimes you have got no choice but to go with a solution because of what it can do. But I think the days of LoadRunner being the only solution out there for this kind of testing have gone. There are some very good competitors now and where the competitor can do the job, you will save money.  

On the scale from one to ten where one is the worst and ten is the best, I would rate Micro Focus LoadRunner Professional as about a six. I have not taken a good look at the latest version, but my current experience with the version of the product we use has not been great.  

The primary use is for load testing and stress testing of applications. It tests whether services are running at an optimal point or not. It can also be used for benchmarking with other industry standards.

My admin was very useful for finding out how the system responds to load, stress, and normal situations, as well as benchmarking with other industry competitors. It also improved our response time, memory delegation, and CPU delegation. In addition, we used LoadRunner to optimize our database and website.

The features that are most valuable are synchronization capabilities and the ability to interact with it on different types of systems and protocols. It is a market leader and it integrates with different applications. When you compare it to JMeter, LoadRunner is more flexible and has more capabilities than JMeter to support different technologies.

There is room for improvement of the pilot processing and forwarding results based on the dump analysis. We have a generator, root controller, different agents, and an analyzer, so all of these are very important when it comes to LoadRunner.

LoadRunner also has to create a low-cost version that supports simpler testing and only some of the simple features would be included. This would allow people to use LoadRunner and get support from LoadRunner in terms of application testing. It is a principle that if people will get to use LoadRunner for free, LoadRunner will get more business and major payback.

I would rate it 6 out of 10, because LoadRunner is a vast thing, starting from those scenarios and then getting more into users, putting more points, correlation, load testing, and benchmarking. I have this part and hope it works well with the system. All of this is very complex. It takes a long time to learn the system well.

The stability of the solution is mostly good. The only thing is that we used a distributor to fix the problems when something went wrong at one point. If it was working out well, the results of that thing would still give you wrong results. So the distributor is not really good with problems, but the local setup is good with no problems.

We can scale it up at any point in time because we can get a distributor. We can add variants, controllers, claims, and analyzers, so it will support different types of protocols.

We just created load scenarios and we will load the scenarios to a number of virtual users. For example, we can load around 1,000 users for each scenario. In abnormal conditions, we could have 10,000 users. Also, loading them depends upon the role alias and role controller's configuration.

The size of staff required for deployment and maintenance depends upon what application and how many load scenarios you want to do, as well as the complexity of your application, how much hardware you have, and how much software support you have. All of those things will come into the picture.

They would be different types of users like a read-only user, a write-only user, etc.

We use JMeter and LoadRunner together: When we're stress testing and for service or load testing we use JMeter, while we use LoadRunner for basic load testing. We keep it like that but we can see how that mitigation works and the application is also for internal users, so we consider what it does there and we can do a lot of those things.

We are not actually concerned with the technical support of the LoadRunner. We get support for the LoadRunner through the internet.

We previously used JMeter.

The initial setup is not that difficult. When it comes to distribution systems then it will be somewhat difficult, and you need to have a proper network to work on. Another consideration is that at the distribution level and onward we need to put different controllers, unions, and claims at different locations to have a product as accurate as this.

The deployment took some time because we needed to know what type of load is expected in the future and what the break-even point would be, at which point of load. We needed to know the different scenarios that were expected and plan the load testing and the load batterings and always stick to the plan. We needed load analysis, a load plan, and benchmarking standards. All of these need to be collected. This equipment needs to be connected before we can actually begin testing.

The deployment takes around one month man days based on the application status.

We used a consultant for deployment.

We did some return on investment analysis, by using the POC (proof of concept). We compared other tools with LoadRunner, JMeter, and LoadRunner and we got a generated a scenario testing based on the place of publication and we determined that LoadRunner is good.

I don't know the licensing cost, but I think that you would get a discount for normal usage. I think there are different yearly options for different types of usage. It is not only how many users, but also whether it is shareable or not and other criteria involved in each feature. There are additional fees for the users and hardware linked to the processing.

We evaluated JMeter and LoadRunner together to form the selection process. For some scenarios we can use LoadRunner for specification road testing, while for other types of testing we prefer JMeter.

When you compare other products to LoadRunner, LoadRunner has been in the market for a long time. You could use it to integrate with everything. Also, it can generate an input that we can use in any version, it can improve our performance, and we can put input in and we get a command from LoadRunner for that.

Our primary use case of this solution is to do testing for banks and financial institutions. 

At the moment I like to use the TruClient port, add them and create other versions of TruClient, because TruClient is usually too big to handle heavy loads, so you need to recreate a vamp version for whatever you will need. 

Sometimes I have problems when I want to record a script if the application is a little special, like the login part. I think the TruClient works well and they are developing new things there all the time. So it's getting better. I would like to see an easier way to move things from SoapUI to the same addressed services and postman or something like that. Because the developers will use the other products and then the website and then load them at no cost. So if it was just one step to move from one system to another it would be a great improvement.

The solution is quite stable. I've seen a few bugs of course - they will always be there. That's not a big deal and I just work around it. 

I guess scalability becomes a problem when you use things like TruClients. It eats lots of memory, so usually, I have two or three low generators while I have the one controller. We have had some problems with when we are using a BPM system. In the beginning, it was quite difficult when we changed to this other system and we wanted to connect to the home computer. 

We have about 12 users, mostly testers, working on the solution currently. Then we have two people working in our performance center. 

The technical support is good. Whenever we had a problem, they solved it. 

The initial setup was quite straightforward and it's quite easy to install the standard version. Everything takes less than an hour.

My advice to others would be to look into the application differently. In the next version, I would like to see an easier way to create and use TruClient for development and to move it to another version. They already have this function, but it is not very user-friendly. You cannot do a correlation of things now, because it is on a higher level now. So you have to do a lot of work. 

My rating for this solution is nine out of ten. I've used it a lot, so I guess it's possible to make things better, but it's been around for a while, so I think it's very good.

For me as a user, the flexibility in different protocols and things like that which can be tested with the tool.

We're able to test standalone from individual machines.

The stability on the old versions is good. On the newer versions, the bleeding edge is still being worked on.

It's very scalable. No issues with scalability.

Premium support is great, but before that when we just had general support, it was not all that great. We had issues with trying to get support to call us back on tickets and turnaround time on resolution.

It's not exactly straightforward. Their instructions were not all they could have been, but we still got it installed.

It is very strong in providing load testing solutions. It provides a lot of protocol support and its reporting utility is very good. Also, the virtual user generation, controllers, and analyses are good.

It helped our organization to run successful performance testing cycles and to identify the bottlenecks of the application. It helped the clients run their application successfully and smoothly.

We would like to be able to implement the tool in such a way that load testing can also be done in the cloud.

We've used it for four years.

There was an issue that I faced with different versions of the controller and Performance Center. They should have been of the same version.

I didn't have many stability issues. The tool is very stable as you just need to check what protocols you want to test.

No issues encountered.

Customer support is good, and the turnaround time for requests is very fast. Also, they have created a forum where experts will help you without creating a ticket.

8/10

I used JMeter and Oracle Application Testing Suite as well, but according to our client’s wish we went for this solution.

It is straightforward. You just have to go through the read-me document and installation guide provided along with the software.

We did it in-house.

I would say it is true value for the money.

The license cost is much more compared to other tools available, but it will give you a complete package for load testing your application.

I evaluated JMeter and Oracle Application Testing Suite, but this proved better than those.

If you are looking for customer satisfaction, you can simply go and get it.

I mainly use LoadRunner Professional for performance testing of web applications.

LoadRunner Professional's best features are that it's easy to capture scenarios and create scripts for testing, the testing runs smoothly with small resources, and it's very easy to create reports with a variety of indicators.

LoadRunner Professional's parameter data could be improved.

I've been using LoadRunner Professional for over ten years.

LoadRunner Professional is quite stable.

LoadRunner Professional is scalable for small organizations but struggles with more than twenty or thirty agents.

We've also been using IBM Rational Performance Test but have found that LoadRunner is better than the IBM tools.

The initial setup was quite easy, and the deployment took one to two days.

The fee for LoadRunner Professional is very high - about US$500 per user.

I recommend LoadRunner Professional as it supports many protocols and applications and is very easy to set up and use. I would rate it eight out of ten.

The capabilities and flexibility of the Controller, the ability to monitor the systems under test, and the comprehensive results Analysis which saves a great deal of time.

More guidance on the use of the Tru Client protocol which is used for Web interfaces.

15 years

LoadRunner generally deploys without issues.

Its a complex product and there are usually a few bugs in the point zero releases, it is generally pretty reliable.

Sometimes the controller can crash, this is usually after opening and closing several scenarios without closing the controller. Closing the controller after a couple of scenarios opened will avoid the problem.

Scalability is great, I have personally used LoadRunner with up to 72,000 virtual users, and 35 load generators.

I don't have recent experience with HPE customer service.

I don't have recent experience with HPE tech support.

I started performance testing with LoadRunner and I'm still using it. I have used other tools, but its the best all rounder.

Initial setup is straightforward as LoadRunner is pretty much standalone and just needs connectivity to the system under test.

I generally implement it myself.

The cost depends greatly on the needs of the testing engagement.

The free community licence (permanent for 50 VUsers) available from Version 12 is very useful. For small tests this is sufficient and there is no licence cost to use it.

The community licence is excellent for training as well, you get full functionality for as long as you like.

No

Take a good look at your requirements, LoadRunner is the best for those who need a full blown product, or a diverse range of protocols to be tested. However not everyone needs this. There are other products that do parts of what LoadRunner does just as well and cheaper. Because LoadRunner is so widely used, there are many people with LoadRunner experience available, it will be harder to find experience in an alternative tool. There are cheaper and free alternatives, but these generally take longer to set up and use. Often the software licence costs are far less than the professional services cost for a performance testing engagement. The higher cost of a LoadRunner licence will be offset by time saved by using it.