Black box:

Not based on knowledge of internal design or code, determines the functionality of the system.

White box :

It is based on the internal logic of the application and the code. Makes a coverage of code statements, branches, paths and conditions.

Unit :

It is the smallest scale of the test is based on the functionality of the program modules, functions, procedures, modules, classes, etc.. In some systems also are checked and tested the drivers and architectural design.

Incremental Integration:

When new features are entered into the test system is based on functionality, the unit with other modules and integration with the whole program.

Integration :

It is based on the evidence of connections and communications between different modules. It is essential cliente_servidor system or network.

Functional :

Does the black box functional testing requirements of the application and is usually done by the programmer, however, the functional test is done by testers.

System :

It is a black box test including all components of the system from the hardware documentation.

From end to end:

It is similar to the test system but this involves interaction with other hardware, databases and networks.

Health:

Determines whether the new version of software is well done and if you need a new effort in testing software. For example the new version of a program meets almost all requirements
but destroys the database to read, so it is said that this software is not in a healthy condition.

Regression :

It's a new review of the evidence after he has suffered hardship or a change of time or was the change in the environment in which it operates. Currently
there automated tools that make this type of evidence does not bear too long.

Acceptance :

It is the final test based on the specifications of the user or based on the use of the program by the end user after a period of time.

Load:

Based on applications under heavy loads, often used on websites and servers with large data sets where there are points at which degradation of the system.

Performance or response:

It is one of the final tests and is used to define the requirements and software quality, based on load testing and stress. Includes interviews with the user and programmer.

Stress:

It is a load test and performance or response (performance) based on the functionality of the system under heavy loads, a large number of repetitions, managing large data and too many questions to large databases.

Installation and removal:

Determines the efficiency of the processes that the applications installed and uninstalled the program.

Recovery:

Is the test that evaluates how well the system recovers after locking, hardware failures or other catastrophic problems.

Safety:

Assesses how well the system protects against access, internal or external, unauthorized, this test requires sophisticated techniques and tools.

Compatibility:

Evaluates the performance of the software on different hardware, operating systems, networks.

Exploration:

It is an informal test of the software that is not based on any plan or test case and often the testers of the program to learn to explore all possible applications.

Listing:

It is similar to the proof of exploration but the testers must have sufficient understanding of how the program works before you start this test. Includes meeting with analysts and programmers.

User:

Determines if the user performs satisfactorily in the program.

For comparison:

This test compares the pros and cons of the program with the programs set up with the competition.

Alpha :

Is the test when the application is close to delivery to the user. Generally small changes are made in the design of interfaces. This test is done by users.

Beta :

It is the search for defects in the program. It is usually made by users.

Mutation :

This test is based on the introduction of codes outside the program (defects) to review whether these defects can be detected. Requires high availability of computing resources.