Feature Friday – Discover the Benefits of End-To-End API Performance Testing
As David Copperfield once said, “The magic lives in the performance!” Although we all know David Copperfield to be a magician, not a tester, the same can be said in the testing world, as well! Performance testing is such an important and integral part of testing an application. Measuring the time it takes for your webpage to load or the time it takes for some task to be completed are included in performance testing. However, today, we are here to talk a little bit about the performance testing of APIs. But, wait, there’s more! We’re going to be talking specifically about the performance testing of end-to-end API processes. Here to deliver more information for us today are our good friends Parth and Steve; without further to do, let’s delve in!
Tell us more about end-to-end API performance testing offered by Qyrus and its use cases.
We have seen many people take advantage of our API Process testing service in which we enable users to stitch or chain together API tests in order to make a complete end-to-end flow. This gave users the kind of holistic testing they were looking for. With Qyrus, we can check the functionality of each individual API, but with this we can also check that these APIs work together exactly as they should.
Now, being able to test this process’ performance is a big step forward. Previously we were able to see basic response times for each API call in a process, but now we can get a much more granular view that we were never able to before.
Exactly. We are, for example, able to get values back for the response time of the entire API flow, the response time per API, as well as the status code per API. This is tested over a certain thread count, or users accessing the same resource – or in this case flow – concurrently.
Adding more tools and functionality to Qyrus is a never-ending task here. Going further into process testing, you can learn more about our individual API Performance testing here from another Feature Friday. If interested, here’s two more about web performance profiling and mobile performance profiling.
What is end-to-end API performance testing’s overall impact on the testing process?
In general, the impact we see is a larger amount of test coverage. This is because the user is able to reuse these API Process tests that they created for functional purposes, but instead run them to test performance.
Furthermore, users can now in a literal sense perform performance tests on these processes. Prior, on Qyrus, you had to run performance tests on each functional API in order to get an idea of what the performance metrics were like. But still, you had an incomplete picture.
How might this feature help testers, developers, and business technologists? What value can this feature bring?
Like Steve previously stated, a tester can now actually run performance tests across a whole process or end-to-end flow and can see the bigger picture, you might say.
Beyond that, developers could test max stress levels for their APIs when they’re all connected in a flow and make adjustments from there. This can be helpful at catching performance flaws early on in the development process.
And for business technologists, they can get better information about performance and at a faster pace. These test results can be shared, emailed, or even downloaded.
Collaboration, although not an explicit feature on Qyrus, is a key quality that Qyrus upholds across the platform – whether that be collaboration during the test building phase, collaborating on test executions, or sharing reports and collaborating with other teammates that way. You can learn more about our collaborative efforts in one of our previous Feature Fridays.
How do you see this feature impacting day-to-day operations across organizations?
Well, to start, like we mentioned before, users are now enabled to run these types of tests across entire end-to-end API flows. This would mean that, day-to-day, testers no longer have to run individual process tests over and over again. Now, they can do it all from one place – and that includes functional testing, as well.
Furthermore, no prior coding knowledge is required. It’s all been simplified down to a no-code experience where any user can understand what’s going on. And as a last point, all tests here are just reused from our API Functional testing section. Nothing has to be recreated, just reused and modified if necessary.
What might we expect to see improved about end-to-end API performance testing?
Well, we’d like to see more performance metrics be added for the user’s benefit. Things are still in progress here, and improvements never stop coming to the platform as a whole.
Right. Some of these metrics might include ones that are currently present under our “Advanced Metrics” under API Performance testing. This might include response latencies over time per API, hits per second, transactions per second, active threads over time, and a general summary tab where we can see max, min, and average response times and more.
If the amount of features that we have across the platform isn’t magical, then I don’t know what is! Well, besides David Copperfield. Oh, and Fridays, those sure are magical! Speaking of, it must be Friday, since you’re reading this. Please have an enchanting weekend and enjoy the time off with family and loved ones. Don’t worry, though, next week we’ll pull some more fantastical features out of our hat for you to learn about! See you next week on our next iteration of Feature Friday!