This is an experiment!
With the rise of AI language models and the huge population of ChatGPT i had a question in my mind , how AI can be helpful within the performance testing lifecycle.
So i decided to try Google Bard and ask the following 3 questions.
- Design a load testing workload for e-commerce site with traffic expectation of 1000 users in the first week
- What is the expected response time for an online CMS.
- Create a performance testing execution plan starts with 100 users concurrent and ends with 20000 users concurrent
The answers were way better than i expected , i will share here a snippets from the answers but you can find the full answers links in the end of this article.
The 1st question was to design a workload , the answer includes a number of aspects like the user profiles , the scenarios or the “Request Mix” as the answers and also describe the flows and more.Show the following screen shots.
The 2nd question was about the expected response time or the recommended response times , although the answer was so general and also that the expected response time for any system is a defined requirement but the answer is reasonable , see the following screen shot.
The 3rd question is the more advanced one , i wasn’t expected Bard to design an execution plan but it did , not so perfect but for sure you can tune it , see the answer below.
To be honest , this was a good surprise.
This can and will help a lot of people who are engaging in a performance testing projects for the first time and also will help the experienced testers and save them some time.
I tried the same question on “Bing Co-pilot” but the answers were search results and links.
Hope you find this useful.
Have a good day!
Links
https://g.co/bard/share/fac2218e7d97
The Test Therapist 
Leave a comment