“Lost time is never found again” – Benjamin Franklin
What is Response Time?
After a request is sent to an application, “response time” measures how long it takes for a response to return back from the application. (This is also sometimes referred to as “round-trip” time.) Response time is important in performance testing because it represents how long a user must wait for a request to be processed by an application. Slow response time equals an unhappy user experience, and may also result in the loss of revenue.
Brick and Mortar Example:
To use a real world example: imagine a gas station. Let's say that the gas attendant takes one minute to fill your car's gas tank. If there are cars in front of you waiting for gas it may take you some time before you get to the gas attendant. So, the total “response time” would be the one minute it took to fill your gas tank plus the time you had to wait beforehand.
Let's say that Joe's Gas Station has an online site that you can visit to add funds to a gas card. You start a browser, and navigate to Joe's Gas online and make a request to load the card. The browser submits a request to the website, and you wait to be directed to the Joe's Gas confirmation page in the browser. If the site is handling a large number of customers at the same time, your request may not be instantly processed; just like the brick and mortar example, your request may wait in a line (queue) before it's actually processed. Your “response time” would be the amount of time that elapses between the moment you entered your request and the time you were returned to the HTML screen in the browser.
How is it Measured?
Response Time is measured in a test tool by surrounding an important business process with
Start and End transactions. A business process could be an action or set of actions a user performs in an application in order to complete a business task. Some examples would be a Google search, a login to an application or a book purchase on Amazon.com. To measure a business process within a tool like Loadrunner, you would use the lr_start_transaction to insert immediately before the action you wish to measure, and then add an lr_end_transaction immediately afterward.