Authors
Feature Friday – Best Way to Capture the API Performance Metrics
Hello and welcome to another edition of Feature Friday, where we highlight specific features or even a specific set of features that help better enable testers to thoroughly and precisely test their applications for defects and functionality. However, we do more than just cover functional testing, we also have capabilities to capture performance profiling metrics. Today, we are specifically going to be covering how we can help enable better performance testing from an API perspective. Joining us are Tim and Jorell who’ll give us more information on the topic!
Tell us more about API performance profiling metrics offered by Qyrus and its use cases.
Tim:
Performance profiling metrics on Qyrus enable testers to get a more granular view of how their API or APIs perform over time. Specifically, Qyrus is testing and gathering these metrics from a client-side perspective. Some basic information is given, such as the response time per thread and the resulting status code per thread.
Jorell:
However, there is also some more advanced information given, such as the minimum, average, and maximum response times, the standard deviation of response time, response latencies over time, active threads over time, hits per second, and transactions per second.
Tim:
Specifically, this enables testers to get a better understanding of how their APIs perform from a very critical client-side perspective. Ensuring proper response times and that the APIs can handle larger amounts of requests is important.
What is the overall impact that this tool has on the testing process?
Tim:
This has an impact in terms of test coverage and reporting. From the coverage side of things, all of the performance tests are reused from the functional side of API testing on Qyrus. This allows for a high degree of reusability and a low degree of maintenance.
Jorell:
That also leads to a degree of effort reduction since things are highly reusable. Furthermore, we are looking to add more useful metrics throughout the process. Recently, we have added many more metrics for consumption by the user, such as hits per second, transactions per second, and the min, max, and average response time for the API call.
How might this feature help testers, developers, and business technologists? What value can this feature bring?
Tim:
In terms of developers, they could utilize the performance testing to test out the APIs they had just finished developing. From that perspective, this type of testing can be done earlier than usual and at a higher degree than usual.
Jorell:
Business technologists will find it extremely easy to not only set up these types of tests but also to execute and view the reports of these tests, as well. The whole point is to make things easier, so that business technologists, can just import the already-created API tests from the functional area.
How do you see this feature impacting day-to-day operations across organizations?
Jorell:
I know we’ve said it what seems like a million times, but the fact that we can just reuse tests cannot be understated. Cutting the test building time in half, as well as allowing for much easier, quicker, and simpler executions, would save a lot of time in terms of day-to-day operations.
Tim:
Yeah, and that comes from an individual tester’s standpoint but also from a higher level; overall, everything is sped up. Day-to-day operations on the creation of test scripts and API tests can be cut enormously depending on the amount of testing done on the performance side.
Performance testing is an extremely important aspect of the testing process as a whole. Things may be functional, but if it takes 5 minutes for an API to respond after the click of a button, things are seriously wrong. With Qyrus, all aspects of testing are encompassed on one singular platform. We hope you’ve found this informative and invite you to join us next week for our next Feature Friday!
