Term
| How to select the subsystems for integration testing? |
|
Definition
|
|
Term
| What is integration testing strategy? |
|
Definition
| determine order in which the subsystem are selected for testing and integration |
|
|
Term
| Describe big bang approach for integration testing |
|
Definition
| runs each individual, and combine all tests and run it at the same time |
|
|
Term
| Do you think big bang approach for a 3-layer-design is a good idea? |
|
Definition
| it's hard to detect where the problem is |
|
|
Term
| Describe bottom-up testing strategy |
|
Definition
start from bottom and go to up, with layer above including the layer down tests
Lets the developers do the integration testing as soon as the modules in a single level are ready.
Easier to track testing progress.
Easier to find bugs. |
|
|
Term
| What is the pro for bottom-up testing strategy? |
|
Definition
useful for system like: object oriented system, real-time system, system with strict performance requirement
No stubs needed |
|
|
Term
| What is a special program which is needed to do bottom-up testing? |
|
Definition
|
|
Term
| What are the cons for bottom-up testing? |
|
Definition
| Bad for functionally decomposed systems because the it will run most important subsystem last |
|
|
Term
| What is a good design pattern used for integration testing? |
|
Definition
|
|
Term
| What is a software needed to run bottom-up testing? |
|
Definition
|
|
Term
| Describe top-down testing |
|
Definition
Test the top layer first, then go down later, combining layer above
Top-down Testing makes it easier to find a missing branch link. |
|
|
Term
| What is a software needed to run top-down testing? |
|
Definition
|
|
Term
| What are the pros for top-down testing? |
|
Definition
1. Test cases can be defined in terms of functionality of the system
2. It's fast |
|
|
Term
| What are the cons for top-down testing? |
|
Definition
Writing stubs is needed.
Maybe large number of stubs.
Does not test individual subsystems thoroughly |
|
|
Term
| Describe sandwich testing |
|
Definition
| tests are converged to target layer out of 3 layers |
|
|
Term
| In sandwich testing, how to select the target layer if more than 3 layers? |
|
Definition
| Try to minimize number of stubs and drivers |
|
|
Term
| What are pros of sandwich testing? |
|
Definition
| Top-down and bottom-up can be done parallelly |
|
|
Term
| What are cons of sandwich testing? |
|
Definition
| does not test individual subsystem thoroughly before integration |
|
|
Term
| Describe modified sandwich testing? |
|
Definition
Phase 1: top layer, middle layer, bottom layer runs in parallel
Phase 2: top layer + middle layer, bottom layer + middle layer |
|
|
Term
| What is continuous integration. |
|
Definition
| A software development technique where people integrate their work frequently and automatically test |
|
|
Term
| What is the goal of integration testing? |
|
Definition
| Tests interfaces of subsystems |
|
|
Term
| What are 7 steps in integration testing |
|
Definition
1. component test
2. sun in bigger subsystem
3. test functional req
4. test subsystem decomposition(dependencies)
5. test non-functional requirements
6. log tests
7. repeat 1 to 7 until system is fully tested |
|
|
Term
| What are two types of system testing? What are goals for each? |
|
Definition
Functional testing: test functionality of a system
performance testing: try to violate non-functional requirements |
|
|
Term
| What are 4 types of performance testing? |
|
Definition
1. stress testing
2. volume testing
3. compatibility testing
4. configuration testing |
|
|
Term
| What is the goal for acceptance testing? |
|
Definition
| Demonstrate system if ready for operational use |
|
|
Term
| What are two types of testing needed for acceptance testing? |
|
Definition
|
|
Term
|
Definition
client use software in dev's environment
software used in controlled setting |
|
|
Term
|
Definition
client's environment
maybe many different environments |
|
|
Term
| What are 4 testing steps? |
|
Definition
What needed to be tested
How it needs to be tested
Write test cases
Create test oracle |
|
|
Term
| Under which circumstances to use white vs black box testing? |
|
Definition
White Box: To test the dynamic model, or in other words, to test the state and internal variables of a unit. Ensures thoroughness as every line can be tested (statement, loop, path, and branches).
Black Box: To test the functional model. For x inputs check for y outputs. Build a list of inputs with expected outputs based on both valid and invalid ranges. Need to have a basic understanding of the box.
Black box also used when you don’t have access to the internal systems but need to make sure it’s working wrt your implementation |
|
|
Term
| what is function coverage? |
|
Definition
| Percentage of methods (functions) that were included in test cases |
|
|
Term
| what is statement coverage |
|
Definition
|
|
Term
|
Definition
| each branch in if-else executed? (if part and else part) |
|
|
Term
|
Definition
|
|
Term
| what's condition coverage? |
|
Definition
| percentage of conditions for true and false |
|
|
