Investigating Online Application Performance with Login PI
As many companies do, we use a CRM system.
Recently, I have been getting complaints about our cloud CRM system being slow. I tried to investigate this by shadowing one of our users to see what was wrong. As expected, everything was fast. 15 minutes later, The same user reported slowness again. How could I investigate this without bothering the users?
2020 UPDATE: Login PI is now Login Enterprise
We’ve put together this brief FAQ to help answer some of your questions.
Using Login PI to Investigate Performance Issues
As it turns out, I can use Login PI for this. A colleague and I set out to create a workload that would log in to the CRM system (Microsoft Dynamics) and do actions a real user would. For example, look up accounts and leads, browse opportunities, navigate through different sections and generate documents.
This “workload” we created mimics our real users and sends us back valuable data we can use to investigate what actions are slow at certain times and when.
The screenshot you see above is actual data from the measurements Login PI did. As you can see, I am able to drill down on what action took longer than normal to complete, and when it took longer to complete.
So, how do I create the workloads? There are two options to do so. The first option is to create the script manually. The second option is to use our Application X-Ray tool that comes with our scripting toolkit, which you can download from the Login PI 3 Virtual appliance. To read more about the Application X-Ray, click here.
We can use the Application X-Ray tool to select certain controls and content on the screen and interact with it. The engine we use can detect when buttons and other things appear on the screen, we can use this to our advantage to navigate through the system.
For example, we open a menu that slides out, then detect if the “My Apps” button has appeared on the screen. We gave this a 5 second timeout. This means that the engine has 5 seconds to detect if the button is there or not. The code you see below is what we used to do this (obviously it will be different for every use-case).
The continuous monitoring of our CRM system has proven that there are certain times when our platform is performing under par. I now know where to look when slowness occurs and focus on the issues instead of guessing where the problem lies.
This is just one of the many uses Login PI has. If you are experiencing issues in applications and would like to know when these issues occur, and how long they take to complete, Login PI is the application for you!
Ready to maximize your end-user experience?
Start maximizing your end-user experience and make your VDI environments run better and operate smoother.