# Integration testing and performance test design

*MuleSoft Platform Integration Architect Study Guide · Designing automated tests for Mule applications · Last updated November 13, 2026 · https://www.certifyforce.com/study-guide/mulesoft-platform-integration-architect/designing-automated-tests-for-mule-applications/integration-testing-and-performance-test-design*

How to tell unit, integration and functional test requirements apart, and how to design a Mule performance test: prerequisites and criteria, a controlled environment, consistent tools, component isolation, representative workloads, iterative execution and resource monitoring.

- Certification: Salesforce MuleSoft Platform Integration Architect
- Exam section: Designing automated tests for Mule applications
- Study guide: https://www.certifyforce.com/study-guide/mulesoft-platform-integration-architect

**Official exam objective:**

> Identify test requirements and scenarios that are best addressed using integration testing or performance testing.

## What the exam expects

- Decide whether a requirement calls for a unit, integration or functional test, and what environment and data checks an integration test needs.
- State the prerequisites that must hold before a performance test starts, including the criteria to establish.
- Design a controlled performance test environment for CloudHub and on-premises deployments.
- Isolate components to find a bottleneck, and build representative workloads.
- Run performance tests iteratively and monitor the resources Mule consumes while they run.

**Integration test:** validates that different units of code and modules work together as intended. **Functional test:** tests the use cases envisioned for the application as a whole. **Performance test criteria:** the expected workloads, the specific need (throughput, response time, concurrency or large messages) and the minimum throughput and maximum response time. **Representative workload:** a test scenario built from real user behavior, workload and payloads. **Think time:** a pause between requests that imitates real users ([Test Mule Applications](https://docs.mulesoft.com/mule-runtime/latest/test-mule-applications), [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations)).

## Key facts

| Fact | Detail | Source |
|---|---|---|
| Integration test goal | Units of code and modules work together as intended | [Test Mule Applications](https://docs.mulesoft.com/mule-runtime/latest/test-mule-applications) |
| Sandbox data | Verify the correct state before and after the integration test | [Test Mule Applications](https://docs.mulesoft.com/mule-runtime/latest/test-mule-applications) |
| First prerequisite | The app works as expected; a wrong flow gives false-positive data | [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations) |
| Large-message need | Payload bigger than 1 MB | [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations) |
| CloudHub test placement | Load client and backend on dedicated machines in the worker's region | [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations) |
| On-premises test placement | Load client, runtime and backend in the same Anypoint VPC | [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations) |
| Changing load tools | Execute all tests again | [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations) |
| Tuning cadence | One change at a time, each test run several times | [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations) |
| CloudHub tuning scope | Mule flows and configurations | [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations) |
| CloudHub monitoring | Runtime Manager dashboard or Anypoint Monitoring | [Performance Monitoring](https://docs.mulesoft.com/mule-runtime/latest/tuning-monitoring) |

## Check yourself

- When does a performance use case count as "managing large messages"?
- Why parameterize outbound database and HTTP addresses with placeholders?
- Which two tools does the guidance name for monitoring Mule app performance on CloudHub?
- Which `sar` command and metrics does the guidance use to watch memory on Linux during a test?

## Sources

- [Test Mule Applications](https://docs.mulesoft.com/mule-runtime/latest/test-mule-applications)
- [Performance Test Validations](https://docs.mulesoft.com/mule-runtime/latest/tuning-test-validations)
- [Performance Tuning Guide](https://docs.mulesoft.com/mule-runtime/latest/tuning-guide)
- [Performance Monitoring](https://docs.mulesoft.com/mule-runtime/latest/tuning-monitoring)
- [Profiling Mule](https://docs.mulesoft.com/mule-runtime/latest/profiling-mule)
- [MUnit Test Structure Fundamentals](https://docs.mulesoft.com/munit/latest/munit-test-concept)
- [MUnit Overview](https://docs.mulesoft.com/munit/latest/)

## About the full unit

The "Core concepts", "Exam traps" and "Worked scenario" sections of this unit, its 6 linked practice questions, and unlimited timed mock exams are included with the MuleSoft Platform Integration Architect pack (USD $19): https://www.certifyforce.com/practice-exams