Please enable JavaScript.

Coggle requires JavaScript to display documents.

QA (Performance Testing (Measurements (Performance testing process (Design…

- - - - Time taken by the application to respond to the request, sum of the following
        
        Request time
        
        Depends on network latency
        
        LESS IS BETTER
        
        Processing time
        
        Depends on the complexity of the request or server hardware
        
        Response time
        
        Depends on the network latency
        
        LESS IS BETTER
    - - 1 Way
        
        Number of transactions (request / response) / unit of time [milliseconds / seconds]
        
        MORE IS BETTER
      - 2nd way
        
        (k)bytes sent or received / unit of time [milliseconds or seconds]
        
        MORE IS BETTER
    - - Number of errors / Number of requests
        
        LESS IS BETTER
    - - Not about measure, how well the system expands its capacity in terms of response time, throughput, percentage of errors
        when additional resource are added
    - - The average / maximum response time should be __
      - The system should be able to support __ pages per second
      - The system should be capable of supporting at least __ users per hour
    - - Design and build tests
      - Prepare the test environment
      - Run the tests
      - Analyze the results
      - Optimize
      - Retest
  - - - 2 ways to generate load
        
        Increase the number of users, that access the application in the same time
        
        Increase the number of requests, that the application has to handle independent of the number of users
    - - Smoke test
        
        Test with light load, usually one user to verify the test
      - Load Test
        
        Different load levels
      - Stress Test
        
        Test application beyond its normal users to see upto which point it stays stable and responsive, to see at what point application stops working
      - Spike test
        
        App is subject to brief period of sudden increment in the load beyond its maximum capacity to see if the application is robust enough to work correctly during and after the spike
      - Endurance test
        
        Where the application is subjected to load within its limit for a long duration (hours or days) to see if application has memory leaks or doesn't properly close database network connections