Security/penetration testing is testing how well the system is protected against unauthorized internal or external access, or willful damage. This type of testing usually requires sophisticated testing techniques.

Parallel/audit testing is testing where the user reconciles the output of the new system to the output of the current system to verify the new system performs the operations correctly. Let us say, for example, the currently software is in the mainframe system which calculates the interest rate. The company wants to change this mainframe system to web-based application. While testing the new web based application, we need to verify that the web-based application calculates the same interest rate. This is parallel testing.

Similar to system testing, the *macro* end of the test scale is testing a complete application in a situation that mimics real world use, such as interacting with a database, using network communication, or interacting with other hardware, application, or system.

The QA testing process that was followed in my last company where I worked was like this: First of all the Business Requirement Document was prepared as per the client’s requirement (with the muck-up screen shots). Then on the basis of the requirement document, Test Strategy, Test Plans and Test Cases were written in sequential order. Once the Build is made and deployed to the different testing environments where different types of testing were performed to check whether there are any defects.

Basically Load, Stress and Performance Testing are the same. However, Load testing is the test to check the users’ response time of number of users of any one scenario of the application whereas Performance Testing is the test to check the user response time for multiple scenario of the same application.

Exploratory testing: means testing an application without a test plan and test script. In exploring testing test explore the application on the basis on his knowledge. The tester has no knowledge about the application previously. He explores the application like an end user and try to use it. While using the application his main motive is to find the bugs which are in the application.

Specification-driven testing: means to test the functionality of software according to the user requirements. In this, tester inputs multiple data and monitors the outputs from, the test object. In this testing tester evaluate the showstopper bugs which break the major functionality of the application. This type of testing requires test plan and test.

Regression Testing: When changes in the code of the software are made to fix the previous bug. Then testing needs to be perform to ensure that it will not generate a new bug in the application and it works as specified and that it has not negatively impacted any functionality that it offered previously. Regression Testing is important because of following reason:

- That the application works even after the alteration in the code were made.

- The original functionality continues to work as specified even after doing changes in the software application.

- The alteration to the software application has not introduced any new bugs.

Automated testing cannot replace manual testing because of following reason:

- Automated testing cannot replace the analytical skills required to conduct testing. It can only enhance or supplement manual testing,

- Through manual testing, information can be gathered about the application, and real-time decisions can be made about how to enhance a test that would be too difficult to program into an automated script.

- We need a manual tester to analyze and maintain the automated scripts after they can run.