One thing that comes to mind is TruClient. It's great for emulating real user behavior on an application, especially when you need to consider how the application interacts with the browser UI. It's like using a UFT script but directly within the browser.

However, the challenge is that TruClient can be very memory intensive. In most cases, clients' infrastructure simply can't handle the resource requirements for running extensive TruClient tests.

Ideally, there would be a way to address this memory bottleneck. We encountered a situation where a simple request-response test with HTTP wouldn't suffice. The user interaction involved a browser layer that needed to be factored in for realistic testing.

If OpenText could find a way to work around the memory issue, that would be very helpful. Here's why: sending a request and getting a response from the server is one thing. But users also want to consider the additional layer of the browser interacting with the application. 

To make the test as realistic as possible, you need to test it with different browsers. And to do that with multiple browser windows open, you really need a protocol that addresses this capability. It exists, but most clients can't handle the resource requirements for running such tests.

When we have a new application, recording the application is a pretty tough task. We have tried multiple things. We do scripting or try to record with different settings and on different machines. We try to record multiple times, but we do not know why it is recording and why it is not recording. We do the same thing on different machines. It sometimes records, and at other times, it does not. That is one of the major concerns.

Another issue is that while executing the load test, we recently faced an issue related to the mmdrv process. After the test is complete, from the LoadRunner Enterprise perspective, it does not show anything running, whereas, at the backend, we see that the load is getting generated from the load generator. We got to know that it is because of the rogue mmdrv process that is running even after the test is complete. We have raised this issue multiple times. The patches or solutions that they have given are not working. We have raised a ticket for this one as well. 

For each and every release, one of the criteria that we have is that it has to pass performance testing. It does not matter how big or small the changes are. Irrespective of the extent of changes, performance testing is required. We give a lot of value to performance testing. For us, performance testing is not only the test execution. We consider that as only 10% of the end-to-end performance testing. The analysis portion of it is where we want to spend 90% of our time. Our scripting effort is not too much, but it is still considerable. For a given application, if end-to-end testing takes 16 hours, then out of that, we need to spend at least four to five hours on scripting, customization, test data setup, etc. For the rest of the things, we have to spend about 12 hours. LoadRunner Enterprise can provide value by enabling us to do these analysis and performance engineering aspects within the tool. LoadRunner has a new feature in the cloud for single transaction analysis or single-user testing analysis. It shows all the insights in terms of the response and hot spots. If they can expand that to LoadRunner Enterprise as well, and not just to the cloud, it would be helpful. 

One more issue is that some of the features are enabled in certain LoadRunner versions. They are not available in other versions. To get a specific feature, the problem is that I either have to upgrade to a different tool or completely get rid of it and then use another tool. Instead of that, they can enable the features irrespective of the platform. Each and every tool that they are producing has its own advantages as well as improvement areas. For example, LoadRunner Professional has the advantage of using chaos engineering tools, which is not there in LoadRunner Enterprise. Irrespective of the tool version I am using, I should have the same advantage. I would like to see chaos engineering in LoadRunner Enterprise.

Their support should be more efficient. I do not raise a ticket just to see the answers that are already available on the Internet.

It would be beneficial if LoadRunner could optimize resource usage, especially for protocols that require significant resources, like TrueClient, which interacts directly with the UI. 

If they could improve resource usage, like ingest or for the load generator, using less CPU or RAM memory, that would be great. That's where I have problems. 

In real time, when they ask for 5,000 or 10,000 concurrent users, I have to provision a lot of virtual machines to define this load.

Then there are situations with certain platforms, especially document management platforms, where the technology is so weird that normal LoadRunner protocols cannot detect it. 

So, in that case, I have to use that special TruClient protocol. I have to use the TruClient protocol, which actually clicks on the object. Despite the SQL technology, I can still create a script and test for performance. 

So what I would appreciate a lot is if this protocol would require less resources on a normal virtual machine. I can use fewer concurrent users with TruClient protocols as opposed to almost one hundred with HTTP/HTML. As opposed to many more with HTTP/HTML from, let's say, JMeter. So, optimization at that level for resource consumption by OpenText would be much appreciated.

After they get over the acquisition, the first improvement is going to be tailoring it for their existing stack of other products. How would LoadRunner work for Documentum? How would it work for Business Network? How would it work for other apps? They can have a pre-package or a guide because they are all in the same family as opposed to being outside. When I spoke to the LoadRunner guys, I said, "You are not a part of the family. You have a simple content server test pack, and you have some accelerators that will get me going with your LoadRunner on another OpenText product." They said, "We just started building those because we have just formally become part of the family." That is going to come out. That is going to add value as I roll out LoadRunner and put it on the mix for other customers and new prospects. When those accelerators are there, I would trust them more. I trusted it previously when they were outside or not inside the box. Now that they are both inside the box, I am expecting synergy to be there.

The debugging feature needs to include graphs. 

I'd like to be able to carry out transaction comparisons with previous tests. It's a feature of NetStorm that I'd like to see in LRE. Currently, we can only compare averages to transition response time. Having that ability to compare would highlight any patterns, comparing them with results at regular intervals. We'd be able to know what was going on throughout the testing process. 

Some of the scripts we use require some custom JARS. It requires importing something to the script if those have been embedded. If that process could be automated, it would make a difference. Whether it's MQ, Kafka or JDBC, those kinds of binaries could be part of a bundle. 

The UI and how they show different methodologies of monitoring need improvement. We'd like to integrate data from different clouds as well as different servers. 

Integration can be tricky during the setup process. 

Technical support in not helpful or responsive. 

They have been doing a lot actually to improve the product right now. 

The tool must improve the ease of gathering and reporting when larger datasets are generated. It must provide better and more direct information. There's a lot of digging around to try and address issues when they arise. Anytime we bring up a window, and it's waiting to do something, it must give a visual indication that it's doing something.

The last few years I haven't seen the need for any improvements to the tool other than fancy stuff or maybe moving from client-server to PC or LoadRunner enterprise. There are a lot of tools in the market nowadays where you don't need to spend much time on scripting. Focus on improving the scripless capabilities.

There is not much value added when upgrading/migrating from older versions to newer versions. OpenText should focus on important features and make end-users like us go for Loadrunner/OpenText products as soon as they are in the market. 

Some areas for improvement in the solution are mainly based on the fact that right now, in the market, there are emerging low-code or no-code platforms, like NeoLoad. In Micro Focus LoadRunner Enterprise, we have to work on scripting. So, I need a skillful resource if I want to hire anyone. In the case of NeoLoad, a low-code environment, and platform, I don't have to search for a skilled resource. I can provide cross-skill training to the existing workforce, and they can work for me. In Micro Focus LoadRunner Enterprise, I need to spend a lot of time training people, while on other low-code or no-code platforms, I need not invest that much time.

Offering a direct integration feature would ensure a completely smooth experience. Until now, we had to use different tools running in the back to get a perfectly detailed analysis.

The price of this solution could be less expensive. However, this category of solutions is expensive.

While they keep on working on improving the tool level, the tool has instability in terms of loading and processing. There are a lot of times hanging issues up until version 11 or 11.2. After that, the productions are pretty stable. 

They changed the GUI. They had wanted to change the GUI to improve the look and feel. However, since that time, we see a lot of hanging issues.

The implementation process can be a bit complex. Even with 15 years of experience, I have trouble finding things. 

They need to make scalability functionality better.

We've raised hundreds of bugs in the Ajax TruClient protocol. I don't know how they are doing testing. However, they should test it in other scenarios. When you're releasing a protocol, the test coverage should be more. They need to do a better job at releasing a full product instead of releasing something half-done and fixing it along the way.

The solution is expensive. 

I work more on the administration side. I am not a daily user, but especially during this conference, we are seeing that everything is going to the cloud. We are trying to see what are some of the benefits of going to the cloud.

I have seen some users report some issues, but I have personally not had any issues.

Some minor issues, like controller crashes and script generation, take time. Sometimes, the code is not generated when we record the scripts in the backend. After clearing the cache, the solution sorts out issues, like identifying the root cause.

LoadRunner integrates with specific tools seamlessly. We do have plans to integrate with Dynatrace or Splunk.

The process of upgrading LoadRunner can be difficult and time-consuming.

The solution should increase the free users limit to 500 for all Micro Focus products.

Honestly, there really isn't any area for improvement. I think it's a great product.

Maybe the scroll bars could be a little bit bigger.

The solution is expensive. 

If they had an easy integration with, let's say, New Relic or Dynatrace, that would be something interesting. If we can see server monitoring data in the LoadRunner report, that would be ideal.

A room for improvement in Micro Focus LoadRunner Enterprise is that it should take multiple exhibitions for a particular scenario and have automatic trending for that. This will be a very useful feature that lets users look into how many exhibitions happened for the scenario and their performance, and you should be able to see the data within the Performance Center dashboard.

For example, there's one scenario I'm focusing on multiple times in a month, and if I check five times, there's no way for me to see the trend and find out how it went with those five exhibitions. It would be great if the Performance Center has a view of all five exhibitions, particularly transaction by transaction, and how they happened. If Micro Focus LoadRunner Enterprise shows you the time trends, information about one exhibition to another, and how each performed, it'll be an immense feature, and that should be visible to every user.

Reporting should be simpler in Micro Focus LoadRunner Enterprise. If I did a scenario with one exhibition now, and I did that scenario again, then I should be able to schedule that scenario for the exhibition, and if that scenario is executed multiple times, there should be the option to turn it into a single view that shows you all the transactions, how the performance was, what the trend graph is for a particular time, etc.

The report from Micro Focus LoadRunner Enterprise would show you the difference between two exhibitions, if I did one today at 12:00 PM and another at 12:00 PM tomorrow, but if you want to see the difference between three or more exhibitions, the solution doesn't have that option. To see the difference, you'll need to do more work in terms of uploading files and doing the comparisons manually, and this should be improved.

An added feature I'd like to see in Micro Focus LoadRunner Enterprise is a converter. I'd also like a performance file extraction feature in the scripting alone. For example, if I'm using the JMX file, I should be able to convert it within the solution, same with other files such as the HAR and PCAP files. Whatever performance file is there, if I can extract it and make a script, that would be a very valuable addition to Micro Focus LoadRunner Enterprise. Another example is if you're not able to record the script in the solution, if there is that option with a PCAP or HAR file, for example, a converter, that will add some value. There's a conversion for the HAR file, but with the PCAP file, I'm not so sure.

Right now, we are in research mode, and we are yet to adopt the solution. The installation has not been straightforward, and we have had so many problems. We have had to re-install, try to install on a different machine, etc. We have not been able to launch the LRE server itself yet.

It's not a consistent solution. Sometimes, it's executes well, and at other times, graphics will not show up, or we'll need to restart the services, for example.

If I change my host controller, then my graphical report goes missing. I'd like to see this improved so that the graphical report is brought to the analysis. Only the LRE server is codable to give the HTML report.

Also, instead of uploading the script, it would be good to have a check-in/checkout option. At present, because the script is uploaded, the version control is missing. Version control sessions would be nice to have as well.

The solution can be improved by making it more user-friendly, and by including autocorrelation capability.

The solution can be improved by adding more AI to the True Client feature.

The cost of the solution is high and can be improved.

It would be good if we could look forward at the future technology needs we have. I would like to see Micro Focus provide more customer awareness around how LoadRunner can fulfill requirements with Big Data use cases, for example, where you do performance testing at the scale of data lakes. That also applies to when we need to deal with applications that are adopting the latest technologies, where our company doesn't have a footprint. It would help us to have a better view and be prepared to address those requirements efficiently.

The Micro Focus team has done a good job of introducing us to product owners and product managers, and in talking about the upcoming roadmap and features of the tool. That's been quite good. But when it comes to technologies our company has yet to adopt, I would like to see an indication from Micro Focus of how one does performance testing and what kinds of challenges can we foresee. Those kinds of studies would really help us.

The reporting has room for improvement.

Micro Focus's technical support could be more responsive.

Something that is missing is a platform where I can share practices with my team. I would like to be able to inform my team members of specific best practices, but at this point, I can only share scripts and stuff like that with them. Having a private community for my own team, where I can share information about best practices and skills, would be helpful.

OpenText needs to improve in terms of support. With the same support plan but when the product was owned by HP, support was more responsive and better coordinated.

The monitoring and related analytical capabilities for load tests should be brought up to industry standards. This product integrates well with tools like Dynatrace and AppDynamics but having the built-in functionality improved would be a nice thing to have.

We'd like the product to include protocol identifiers whenever a tester wants to test a new application. They need to at least capture traffic and analyze and send a recommended protocol. 

They should give insights (passed/failed) with throughput achievable of the load test based on the throughput and errors in our previous tests without using the trend option for load tests. 

I think better support for cloud-based load generators would help. For example, integrate with Amazon AWS so you can quickly spin up a load generator in the cloud, use it, spin it down. That way, you're only paying for the actual time that it's being used. I know they have some functionality with that right now, but it could be improved because right now, our projects have to bare the cost of that infrastructure, whether it's an AWS or whether it's their own VMs, so that would help out with that.

Micro Focus LoadRunner Enterprise needs to add more features for Citrix performance-based applications testing. This was one of the challenges we observed. Additionally, we experienced some APIs challenges.

It needs integration with open-source tools. It should also have version control of scripts, SDI, or GitHub. I'd also like to see some kind of EDI integration and open-source support.

OpenText LoadRunner Enterprise doesn't support some mainframe protocols. We had to build scripts to access the interface. 

It has to be fully integrated into pipelines, it needs to be DevOps friendly. It needs to be easily digestible by management, and certainly developers. It's a developers' world, as it should be. They're the ones who create the applications and solve the problems in those applications. So it has to be positioned to be something that allows a team to make better decisions, to move through that progress I mentioned before, from reactive to proactive to predictive. Once you get the predictive you can make better decisions on how you should be teasing things, and Performance Center will have to follow the same trajectory. It has value, but the value needs to evolve and mature along with other aspects of application development.

I know there are integrations with continuous testing. It's got tie-ins to some of the newer tools to allow continuous testing. I'd love to see us not have to customize it, but for it to be out of the box.

I have some concern over its foundation for utilizing cloud testing hosts in the most integrated fashion. For example there is reliance in AWS to utilize default VPC, and also there is not deep knowledge about utilizing *nix hosts though they are supported.

With Performance Center, the version upgrade is easy. You just have to roll out the new patch or the new version. It is much easier. I'm not really the right person to say, because I run the environment. We have a specialized team that does development.

OpenText LoadRunner Enterprise needs to improve reporting. 

One thing that always fails at our company is that after you have checked in an application then it usually crashes in some way. You get some strange error message. We found out you can open the test you have set up and usually, it works without the error the second time. So you just close the application test and open it again, and then it is okay. So that is quite confusing if you are new to the product, but you do not care about the inconvenience or even notice it after using the tool for a while. It does not seem very professional and it is really a buggy behavior that should be fixed.

One feature I would like to see included in the next release of Performance Center would be to be able to run more fluidly with True Client so you could put more virtual users in Performance Center. That would help. I'm not sure how easy it is to compile something like that, but it would be valuable.

Almost all the areas improve drastically with each version; however, the correlation of scripts, analysis and reporting can be further improved. Their technical support could also be improved. Recording of the latest applications is an area for continuous improvement.

We encountered cookie banner issues when opening LRE.

Integration with other APM tools like DynaTrace or AppDynamics is a little tedious. Plus, support around cloud solutions and architectures needed to be improved.

I'd like to see better reporting, ultimately, and analysis. I'd like to see the analysis more accurately pin-point where these issues are; a little more feedback immediately in terms of the anomalies that we're seeing during the testing, so we can get alerted much quicker, instead of blowing a whole bunch of our day to test data during a test.

The reporting could be a lot better. It's cumbersome. The VuGen takes a long time to load up, and edit and execute things off those. So a lot of the basic infrastructure.

One of the things we were looking for is more of a DevOps support, like BlazeMeter has. It would be an ideal scenario to incorporate those kinds of features. I know there are some open software products which have that but it would be ideal to see those features in the product.

Canned reports are always a challenge and a question with customers because customers want to see sexy reports. They want to be able to show something to the CIO. So I think the dashboards are one of the features I'd like to see most.

I think it's more of getting into the world where you've got tableau and dashboarding. I think that reporting needs to be a little bit more fancy, as people expect the sexier reporting. They don't expect just to have, "I ran a test. The test ran for this long." I think the consumer's expectations for what reporting looks like have changed a lot. You do an Excel report or a Word report versus, "No, it needs to be a very pretty dashboard."

The product itself, I think it's pretty good. I can't think of anything off the top of my head.

I don't know if it is available or it depends on the licenses, but reports from Performance Center need to be improved. If we can improve the reporting structure and we can view the report on a smartphone, that would be huge.

We always have to get connected if there is a report. It's doesn't need to be the full tool, but just like how HPE Discover has an app, if we can have an app just for reports, that would help.

The migration from LoadRunner to Performance Center was not as smooth as I would have liked. Some of it was not the tool but us, but it still could have been smoother and taken a little bit less time.

It's expensive, so could be cheaper. Also, it should have third-party tool integration.

To improve the product, I think the integrations could be a little bit more slick. It does handle a lot of great integrations, but then some of them can be a little bit clunky to implement. The integration with third-party tools needs to be stepped up a little bit.  

As far as other things that need to be added, it has changed quite a lot recently, and I have not had vast amounts of experience with the latest version. So I am afraid it would not be fair for me to go further in expounding on that question. Things that I talk about may already have been included.  

Mobility, I think, is the biggest for us right now. We're really getting into mobile app testing, native mobile app testing, so native Android and IOS devices.

I think LoadRunner is still getting into grips with me — maybe, I've not used it that much. It's not that popular on the cloud. Also, we have not tried this on mobile platforms with mobile virtual users.

Sometimes the Book Time Slot option hangs if I cancel any test or time slot. Also, if the results file gets corrupted, then we need to contact HP to get the raw results.

I think there is a lot of configuration and customization that can come in for optimizing the recorded load test in terms of CPU and memory used and, accordingly, the service/process that are running this v-user script at load generator. This would be very helpful especially when running a huge number of v-users simultaneously.

Performance Center was a stand-alone server application until its marriage with Application Lifecycle Management in 12.x, when both the architecture and the user interface underwent a major overhaul. As such, it has been rapidly evolving for the past several years as HP worked through the initial UI and performance challenge in terms of a response to improve its stability, then begin to focus on adding features that would facilitate communication of performance data.

Today, they continue to enhance reporting and analysis as the My Performance Center dashboard matures, but the web-based interface is still a little clunky in terms of response to user input, especially in the Workload area where a load test is built. Communication between load testing hosts and components is very complex and sometimes fragile, making manual intervention still too often necessary should an anomaly occur during testing.

LoadRunner Enterprise's reporting should be quicker, easier, and more flexible.

The solution is a very expensive tool when compared with other tools.

The stability in some of the latest versions has not been ideal. They need to work to fix it so that it becomes reliably stable again.

The cloud solution of LoadRunner is not user-friendly when compared to BlazeMeter. They need to improve their cloud offering in order to compete. It also shouldn't be a standalone tool.

I think better or more integration with some of the monitoring tools that we're considering.

We're looking to bring in maybe AppDynamics. I personally don't know the integrations but having good integration tools is going to help us in the future.

I think some of the reporting features could be better. I haven't seen much change in terms of that aspect of it, the report analysis piece. It's been good enough but I haven't seen as much advancement in that space, the reporting analysis.

I'd like to see a more shared data repository. They have their costs up. I think they could expand it out a bit more so we could have it running across synthetic users and protocol types.

I'd like to see feedback from production, to see what scenarios to run, and even better integrations into some other products such as AppDynamics.

The trending reports do not always work as I would expect them to. I have had issues with transaction times being pulled in correctly. 

Additionally, trending a test run can take over 30 minutes when it should take five.

In the DevOps model, performance testing has become a bottleneck. This is because, after the completion of a sprint, people are in a hurry to send it to production but it first needs performance testing. Whenever there is a code change, it takes a lot of time to rescript and debug the script.

The TruClient protocol works well but it takes a lot of memory to run those tests, which is something that can be improved. Basically, it is too resource-intensive.

Performance testing needs to be better integrated into an agile framework.

There should be a way of automatically increasing the load generators on the cloud, without specifically having to spin up the agent and configure it. There is a third-party tool that does this. 

Browser compatibility. IE is no longer the only browser out there, so providing users the choice of what browser to use would be great!

I think Stormrunner is the best example of a cloud based solution, so it may be that Performance Center can have a better, richer, and faster UI.

The UI should be improved in the next version.

I believe the data that demonstrates the automated correlations should be corrected. In addition, because it requires a lot of hardware, true client-based scripting should be improved. That needs to be improved.

Features in the Web UI should be enhanced. We should be able to do everything from the Web UI. without ever needing to download the old ActiveX ALM UI.

It needs to be able to inform us what our current situation is, that is, what version we're on, etc. This is important because we do reinstalls and upgrades once a year, so that information is still valid for us each time we do that.

• Monitoring tools which are available in Performance Center.
• More real-time monitoring should be available for the system under test. 
• Dashboard creation should be implemented, so we can get the results in a desired format.

Deletion of collating test results by admin access only.

Storing the results of our tests requires the use of a lot of memory in our data center.

It would be nice if the Citrix protocol could be updated.

• Backend integration
• Performance
• UI improvement
• Integration with third-party tools

Some features could sometimes disappear without reason. It's extremely recommended to take time to evaluates newer versions before upgrading.

I think some enhanced support, be it online or documentation, would be appreciated.

Mobile testing - integrate it with a mobile UI to automate scripts and run it concurrently.

• New features have been added in latest version and need to be improved with the DevOps integration.
• We are expecting more flexible to use Jenkins in continuous integration going forward. 

More features to support testing in DevOps environment(s).