Thanks to those of you who attended our last webinar, How To Combine Front-End and Back-End Testing, featuring our Chief Technology Evangelist Michael Sage and BlazeMeter's VP of Customer Success, Ophir Prusak.
Everyone knows that front-end testing is crucial to make sure your web and mobile apps are meeting the needs of your users and customers. But how do you know what will happen to your front end when your web or mobile app is under heavy load?
Michael and Ophir set out to help answer this question, plus more. Together, they covered many valuable topics, including:
The fundamentals of approaching performance vs. front-end testing
Step-by-step instructions on getting real-world results from your front-end while applying load to the back-end
Critical issues you need to know about performance testing
They also showed a real-world test in real time using JMeter and Selenium.
Missed the webinar? You can watch it in its entirety below.
Below you'll find the top Q&A's post-presentation:
Q: Why do we need JMeter? Why can’t we do performance testing with Selenium browsers?
A: [Ophir] Great question! The answer is really all to do with scalability. When you’re doing a test with Selenium, you’re usually using real browsers –and real browsers are very resource intensive. On the other hand, when you’re doing a test in JMeter by using virtual users, I can exponentially get a lot more users per machine. Just to give you an idea of ballpark numbers, on a single low-end Amazon EC2 machine, I can get 3,4 or 5 virtual users using browsers in parallel until I hit a bottleneck on the machine. If I’m doing it with a JMeter, I can get 1,000 people in parallel. So you’re looking at around 250X more users when I’m doing a JMeter test per machine than Selenium. So if you’re looking at ten or 100 users, you can do it with Selenium but when you’re looking at tens of thousands of users, it just won’t be able to support it.
Q: Can I test behind the firewall?
A: [Michael] Yeah, absolutely We have a utility called ‘Sauce Connect’, which creates an encrypted tunnel between your environment inside your firewall or your DMZ and our grid and you get a dedicated virtual machine to act as the tunnel property. You run this utility and it appears as if the Sauce Labs grid is inside your network and it’s all done in an encrypted fashion.
[Ophir] At BlazeMeter, we also have the ability to run behind the firewall. It’s a different type of solution. It does require having something we call a BlazeMeter agent, which is basically a load generator which sits behind the firewall, which you can still control through your browser but you have a local machine which is creating the requests for you.
Lastly, please follow our friends at BlazeMeter at @blazemeter, Sauce Labs at @saucelabs, Michel Sage at @mondosage, and Ophir Prusak at @prutwo to keep up with the latest. Feel free to share this webinar using the hashtag #frontandbackendtesting.
