When creating your performance test, use your current data to recreate your environment. Make it as identical to your application as possible.

This process will ensure that any performance tests conducted in different environments will come to the same conclusions as they would on a "live" application server.

Users are your most valuable assets. Collecting real datapoints from them, as well as their interaction trends, will help you extract the most value out of your performance tests.

Be sure to pay extra attention to these details:

• The daily and hourly average transaction volume
• Key business flows for users and their volume break up
• Key transactions and actions taken by the users
• The frequency of user check-in per week/month/year
• Geographies from which clients are trying to connect
• The average "think time" that users take while moving from one page to another
• The average pacing time
• The "hit-ratio", which takes into consideration the number of hits and misses in a day
• The time of the day/week/ month when the application is under a peak load
• The corresponding peak load data

Click here to go to the next page.