Novatesting has issued the article named “the Levels of Software Testing – Overview” discussing the four levels of software testing, of which levels, the first level is thoroughly looked into in the article “Unit Testing – A Vital Part of Software Development”. Today, the second level of those, Integration testing, is going to be worked out below.

1. What is Integration testing?

Integration testing is an important part of software development that aims to verify the interactions between different components and systems. This type of testing ensures that all the individual parts of the software work together seamlessly and efficiently. Integration testing helps identify any issues or compatibility problems early in the development cycle, which can save time and resources in the long run.

Hence, integration testing is also known as I&T (Integration and Testing), String Testing and sometimes  Thread Testing.

2. Why is Integration Testing Important?

Integration testing plays a crucial role in the software development process because it helps ensure that the various components of a software application work together as intended. It identifies any gaps or issues in the communication between different parts of the software and helps resolve them. This helps to catch and fix any problems early on, before they become major issues that are more difficult and expensive to fix.

Integration testing also helps to verify the functionality of the software as a whole. It makes sure that all the components are functioning properly and can be integrated into the final product without any major issues. This helps to improve the overall quality and reliability of the software.

Even though each module has been Unit Tested, the error still exists for some reason like:

• Because each module is designed by an independent developer, with different programming logic and knowledge, there may be errors when integrating modules together.
• Customers will change design requirements during module development (adding requirements, updating requirements when they find it unreasonable…) and these new requirements may not be subjected to unit testing or integration, errors will arise.
• The interfaces of the modules in the software with the database may not be compatible.
• When integrating modules into the system may not be compatible with the general configuration of the system.
• Incomplete exception handling can cause errors.

The Test Case of Integration Testing is different from other Test Cases, integration testing focuses mainly on interfaces and the flow of data or information between modules. Since unit testing has been tested for each module, there is no need to re-test here.

3. An example test case of Integration Testing:

Test Case Title: Verify Payment Gateway Integration

Objective: To verify that the payment gateway integration is working as expected and that payments are processed correctly.

Test Steps:

    Navigate to the e-commerce application’s checkout page.

    Enter the required information, including the product details and payment information.

    Click on the “Place Order” button.

    Verify that the payment gateway interface is displayed and the payment information is correct.

    Complete the payment process.

    Check the order confirmation page to verify that the payment was processed successfully.

    Verify that the payment information is reflected in the order history and in the database.

Expected Results:

    The payment gateway interface should be displayed correctly and the payment information should be correct.

    The payment should be processed successfully.

    The order confirmation page should display the payment was processed successfully.

    The payment information should be reflected in the order history and in the database.

4. What are the Types of Integration Testing?

There are several types of integration testing, including:

(1) Big-Bang Integration Testing: This type of integration testing involves testing all the components of the software at once.

(2) Top-Down Integration Testing: This type of integration testing starts with testing the highest level of the software and then moves downwards to the lower levels.

(3) Bottom-Up Integration Testing: This type of integration testing starts with testing the lowest level of the software and then moves upwards to the higher levels.

(4) Incremental Integration Testing: This type of integration testing involves testing small groups of components at a time and then integrating them one by one.

(5) Hybrid/Sandwich Integration: The sandwich/hybrid strategy is a combination of Top Down and Bottom up methods.

4.1. Big-Bang Integration Testing

All components are integrated at the same time, then testing is carried out.

Advantages: Convenient for small systems.

Defects:

• Difficulty in detecting bugs.
• Given the number of interfaces that need to be tested by this method, some of the associated interfaces that need to be tested can easily be overlooked.
• Since Integration testing can only start after all the modules are designed, the testing team will have less execution time during the testing phase.
• Since all modules are tested concurrently, high-risk critical modules are not isolated and are prioritized for testing. Modules that are related to the user interface are also not isolated and are prioritized for testing.

4.2. Top-Down Integration Testing

This type of integration testing starts with testing the highest level of the software and then moves downwards to the lower levels.

Advantages:

• Early Detection of Issues: By testing the higher-level components first, issues can be detected and resolved early in the development cycle, saving time and resources.
• Clear Focus: By focusing on the higher-level components first, the testing process can be streamlined and more focused.
• Improved Test Coverage: Top-Down Integration Testing ensures that the higher-level components are thoroughly tested, providing a better understanding of the software system as a whole.
• Reduced Development Costs: By detecting and resolving issues early in the development cycle, Top-Down Integration Testing can reduce development costs and help to ensure that the final product is of high quality.

Defects:

• Limited Test Coverage: Because the lower-level components are tested later in the process, there may be limited test coverage for these components.
• Dependencies: The testing process may be impacted by the dependencies between components, making it difficult to test the higher-level components until the lower-level components are complete.
• Complexity: Top-Down Integration Testing can be complex, especially for large and complex software systems, as it requires a thorough understanding of the software architecture and interactions between components.

4.3. Bottom-Up Integration Testing:

This type of integration testing starts with testing the lowest level of the software and then moves upwards to the higher levels.

Advantages:

• Better Test Coverage: By starting with the lower-level components, Bottom-Up Integration Testing provides a more comprehensive test coverage, including the interactions between components.
• Simpler Testing: Bottom-Up Integration Testing is often simpler and more straightforward than Top-Down Integration Testing, as it focuses on testing the individual components before integrating them.
• Reduced Dependencies: By testing the lower-level components first, Bottom-Up Integration Testing reduces the dependencies between components, making it easier to test each component individually.
• Early Feedback: Bottom-Up Integration Testing provides early feedback on the functionality of the individual components, which can be used to identify and resolve issues early in the development cycle.

Defects:

• Limited High-Level Testing: Bottom-Up Integration Testing may not provide as much testing coverage for the high-level components as Top-Down Integration Testing.
• More Complex Integration: Integrating the individual components can be more complex in Bottom-Up Integration Testing, as the interactions between components must be taken into account.
• Increased Time and Cost: Bottom-Up Integration Testing can be more time-consuming and costly than Top-Down Integration Testing, as it requires more testing resources to test each component individually.
• Incomplete Testing: Bottom-Up Integration Testing may not cover all the possible interactions between components, leading to incomplete testing and potential issues in the final product.
• Delayed Feedback: Bottom-Up Integration Testing may provide feedback later in the development cycle, making it more difficult to resolve issues and making it more expensive to correct problems.

4.4. Incremental Integration Testing:

This type of integration testing involves testing small groups of components at a time and then integrating them one by one.

Incremental Testing

Incremental approach is in some case implemented by two different methods, Bottom Up and Top Down as described above.

Advantages:

• Early Detection of Issues: Incremental Integration Testing detects issues early in the development cycle, allowing developers to resolve problems before they become more difficult to correct.
• Clear Focus: By testing components incrementally, the testing process is more focused, allowing developers to concentrate on specific areas of the software system.
• Reduced Complexity: Incremental Integration Testing reduces the complexity of the testing process, as the software system is tested in smaller, more manageable pieces.
• Improved Test Coverage: Incremental Integration Testing provides improved test coverage, as it allows developers to test each component thoroughly before integrating it with other components.
• Reduced Time and Cost: Incremental Integration Testing reduces the time and cost of testing, as it is less complex and focuses on testing smaller components.

Defects:

4.5. Hybrid/Sandwich Integration

The top modules are tested at the same time as the lower modules, and the lower modules are integrated with the above modules and tested. This strategy uses Stubs (*) as well as Drivers (**).

Advantages:

• Comprehensive Test Coverage: Hybrid/Sandwich Integration Testing provides comprehensive test coverage, as it combines the strengths of Top-Down and Bottom-Up Integration Testing.
• Improved Feedback: By starting with the high-level components and then testing the lower-level components, Hybrid/Sandwich Integration Testing provides improved feedback on the functionality of the software system.
• Reduced Complexity: By combining Top-Down and Bottom-Up Integration Testing, Hybrid/Sandwich Integration Testing reduces the complexity of the testing process.
• Early Detection of Issues: Hybrid/Sandwich Integration Testing detects issues early in the development cycle, allowing developers to resolve problems before they become more difficult to correct.

Defects:

• Increased Time and Cost: Hybrid/Sandwich Integration Testing can be more time-consuming and costly than other testing techniques, as it requires the integration of components in both directions.
• Dependencies: Hybrid/Sandwich Integration Testing may be impacted by the dependencies between components, making it difficult to test individual components in isolation.
• Complex Integration: Integrating the individual components can be complex in Hybrid/Sandwich Integration Testing, as the interactions between components must be taken into account.
• Fragile Test Cases: Hybrid/Sandwich Integration Testing may result in fragile test cases that break easily when changes are made to the software system.

5. What does an Integration Testing Process look like?

Here is an integration testing process irrespective of software testing strategy:

• Prepare integration test plan.
• Design test scenarios, test cases and test scripts.
• Execute test cases, report bugs if any.
• Follow up & retest test cases with errors.
• Steps 3 and 4 are repeated until the matching test is completed.

6. Brief description of integration testing plan

Integration testing includes the following properties:

• Testing method/approach.
• In and out of the scope of integration testing.
• Roles and responsibilities.
• Prerequisites for integration testing.
• Test environment.
• Risk reduction plan.

7. Integration test start and end criteria

Start and end criteria of integration testing phase in any software development model:

Start criteria

• The component/module has been unit tested.
• All high-priority bugs have been fixed.
• All modules are complete and integrated.
• Integration test plan, test cases, scenarios, documents approved.
• Test environment is set up as required for integration testing.

End criteria

• Successful integration test.
• Executed test cases are logged
• All high priority bugs fixed
• Technical documents are handed over.

8. How to perform integration testing to get the best results?

• Define the scope of integration testing: Determine what components of the software system will be tested, and the level of testing required. This will help to focus the testing process and ensure that all relevant components are integrated and tested.
• Identify the integration approach: Choose the integration testing approach that best fits the needs of the software system. This could be Top-Down, Bottom-Up, Incremental, Hybrid/Sandwich, or another approach.
• Plan the integration testing: Develop a detailed plan for integration testing, including the schedule, resources, and personnel required. This will help to ensure that the testing process is organized and effective.
• Prepare the test environment: Create a test environment that mirrors the production environment as closely as possible. This will help to identify and resolve issues that may arise in the production environment.
• Develop test cases: Develop test cases that reflect the requirements of the software system and cover all relevant components. Test cases should be automated where possible to ensure consistent and accurate results.
• Execute the tests: Run the integration tests and document any issues that are identified. Work with the development team to resolve any problems that are encountered.
• Monitor and evaluate results: Monitor the results of the integration tests and evaluate the performance of the software system. Use this information to make any necessary improvements to the integration testing process.
• Maintain test cases: Regularly maintain and update the test cases to ensure that they remain relevant and effective. This will help to ensure that integration testing remains effective over time.

Integration testing is a crucial step in the software development process that helps to verify the interactions between different components and systems. It helps to identify and resolve any compatibility problems early on, which can save time and resources in the long run. Whether performed manually or through automated testing, integration testing plays a vital role in ensuring the overall quality and reliability of the software.

(*) A stub is called by the module under testing.

(**) A driver calls the module to be tested.