Manual Load testing, It happens only in India
Title would be kicking performance expert in IT industry, What the hell I am talking about, How any one can find resources and human resource to do manual load testing.So don’t be so perturb because I have seen one live example of this manual Load testing and This is true incidence because I was also among those hundred people who has just hit the Sing In button as soon as a bell rang at 4 ‘o’ Clock in the evening.
I was also not sure that in real scenario a project manager would asked for 5 minute by all the people on floor to be ready with username and password and to waiting for the signal to hit sing in button . But this happens and it was very fruitful because system was design to perform smoothly with the user capacity of 35 user. This has shown through log file that near about 31 people has hit the sing in page at the same time among those of 100 people. This was sufficient to come out with conclusion about the performance of simultaneous sing in functionality.System has taken 6-7 seconds in login in place of expected 2-3 seconds. Since I have just listen the word performance and I was just familiar with one google’s standard, if site is taking more than 3 second means it is performing terribly.So in my understanding system was not accomplishing the real need or requirement of performance.
Through this incidence I just want to remind that People in India believes in Jugad methodology and wherever they feel like using this they normally do the same.
So I think Now I can say “This Happens only in India ” and I am proud of it.
Nice anecdote …… I remember doing something similar in a project once. We had to load a file of 1000+ payments and as acting lead, I asked my resources (4-5 in number) to get ready and execute the one file each per resource. Results were fruitful. Thumbs up for ‘Jugaad’ tech.
ya some time we need to expand the limit to meet some conclusion and such activities really bring smile on the face of all at the end of the day…..
Hmm…really, in India only? Interesting article. I’ve seen similar type (though not same) scenarios done in the U.S. The manual load testing approach described here doesn’t seem illogical or extreme either. I’d love to hear about a real life manual load test scenario that actually required a scale in the 100-1000+ user range (meaning you actually need that many human testers to help load the system, and choosing to do it with real people over automated test tools).