Mastering Software Test Architecture: Skills, Types, and Methods

1 Software Product Quality Six Attributes

The software test architect must master a range of quality attributes such as functionality, reliability, portability, and more, illustrated with examples from the Windows Calculator (e.g., suitability, accuracy, interoperability, safety, functional compliance).

2 Test Types

Test types are defined from different angles using the software product quality model, linking each quality attribute to appropriate testing focus.

3 Test Methods

Product test wheel visualizes the relationship between quality attributes and test types, helping ensure comprehensive coverage.

Functional testing methods include:

Single normal‑value input

Single boundary‑value input

Multi‑run sequential execution

Multi‑run interaction

Reliability testing methods cover abnormal‑value input, fault injection, stability testing (the "four‑character" rule: multiple, concurrent, repeated, abnormal), stress testing, and recovery testing.

Performance testing verifies that the system meets its specified capacity and throughput, distinguishing between "new" (creation) and "concurrent" capabilities.

Usability testing focuses on consistency and overall usability of the user interface.

4 Test Design Techniques

The four‑step test design method consists of:

Modeling – selecting appropriate models (flowcharts, input‑output tables, equivalence class tables, factor tables) based on test‑point characteristics.

Designing basic test cases that cover the model.

Adding concrete test data to transform basic cases into full test cases.

Expanding with additional cases based on experience and defect‑prone areas.

Test points are classified into four types (process, parameter, data, combination) to guide the modeling step.