Developing high-quality products involves careful definition and tracking of both functional and non-functional requirements. While performance testing falls under the non-functional testing technique, it is just as valuable and vital as functional testing techniques.
What is performance testing?
Performance testing is like checking if the software can handle lots of people using it without getting crash, slowing, or breaking. In other words, it helps to examine the speed, robustness, reliability, and application size of a piece of software under a given workload. The programs must run fast and smoothly.
Let’s understand the impact or importance of performance testing with an example–
Just assume there is one newly launched website on which proper performance analysis was not done, having 1000 users on the launch day, now because of sudden heavy load, some of the APIs are not behaving properly as in slow response, memory, or data leakage, timeout errors, etc. Will it not affect the functional experience of users? It definitely will!
Avoiding non-functional requirements (Performance Testing) can impact our end users more. Do you think anyone will use such a product again and again if it is not usable? Why will anyone visit the system if it is not secure enough, and why will anyone suggest it to his friends and family if it is not sustainable?
Applications sent to market with poor performance metrics due to nonexistent or poor performance testing are likely to gain a bad reputation and fail to meet expected sales goals.
Thus, performance testing is vital to save a product from crashing badly in the market, and paying attention to NFRs from the beginning is the key to project/product/system success.
According to Dun & Bradstreet, 59% of Fortune 500 companies experience an estimated 1.6 hours of downtime every week. Considering the average Fortune 500 company with a minimum of 10,000 employees is paying $56 per hour, the labor part of downtime costs for such an organization would be $896,000 weekly, translating into more than $46 million per year.
Only a 5-minute downtime of Google.com (19-Aug-13) is estimated to cost the search giant as much as $545,000.
It’s estimated that companies lost sales worth $1100 per second due to a recent Amazon Web Service Outage.
Hence, performance testing is equally important as functional testing.
Now how to conduct performance testing?
Because testers can conduct performance testing with different types of metrics, the process can vary greatly. However, a generic process may look like this:
Three major things to consider during performance testing are:
1) Determine performance criteria
Understand your requirements and customers’ needs correctly. If you don't know what needs to be done, and what impact it has on your users then you are going on the wrong path, and you will not be able to put any value.
2) Test plan
Plan testing to simulate production-like behavior (which might not be 100% possible) and then analyze. A key to this is the preparation of test data, as well as the maintenance of the production environment during testing.
As per my experience, for performance analysis, we can divide our use cases on the bases of our customers like large customers, medium customers, and small customers/businesses.
For large customers, our test plan will be on the higher side in terms of numbers and performed load.
3) Result analysis
As per my experience, in performance testing test execution is easier (comparatively) than analysis of results. You need to follow a constructive approach and dig more into suspicious stuff. If things are not working as expected try all possible ways and work around and then finalize your observations.
Final words:
Realistic tests that provide sufficient analysis depth are vital ingredients of “good” performance tests. It’s not only about simulating large numbers of transactions but anticipating real user scenarios that provide insight into how your product will perform live.
Performance test analysis provides insight into the availability, scalability, and response time of the application under test. By doing regular performance tests, your product can work really well. This means it stays up more, needs less fixing, and people can do more on it. Even though making this investment takes time, it's worth it for your business. People will have a great time using your software product, and they'll want to come back again and again.
