03-by-device-type

Pattern 03: By Device Type

Run the same test suite across multiple coverage buckets simultaneously. Designed for teams optimising for coverage rather than speed.

• Appium implementation uses manufacturer buckets (Samsung, Google, OnePlus, Xiaomi) with tier tags.
• Selenium implementation uses browser tiers (Chrome, Firefox, Edge, Safari).

When to use this pattern

• Your app needs to perform consistently across different hardware tiers
• You have compliance or market coverage requirements (e.g. must support devices down to Android 10)
• You want to catch device-specific rendering bugs, memory issues, or performance regressions

How it works

Full test suite (100 tests)
         │
         ├── Appium Slot 1 → Samsung manufacturer bucket (tier: flagship)
         ├── Appium Slot 2 → Google manufacturer bucket  (tier: mid-range)
         ├── Appium Slot 3 → Samsung manufacturer bucket (tier: budget)
         ├── Appium Slot 4 → Samsung manufacturer bucket (tier: legacy)
         │
         └── Selenium Slot set → browser tiers (chrome / firefox / edge / safari)

All slots run identical tests for their matrix target.
Execution ratio impact: MEDIUM — same tests across multiple coverage dimensions.

Tradeoffs

Advantage	Disadvantage
Catches hardware-specific failures	Longer total pipeline (same tests × N devices)
Clear coverage matrix for stakeholders	Flaky tests appear on all devices
Required for regulated markets (RBI, HIPAA device diversity)	Higher pCloudy slot consumption

Current matrix in this repo

Layer	Buckets used
Appium + TestNG	Manufacturer buckets: Samsung, Google, OnePlus, Xiaomi
Selenium	Browser tiers: chrome, firefox, edge, safari

---

Implementation

• Appium + TestNG (Java)
• Selenium (Java — web/browser)