Metrics for Architects
summary
Covers concrete metrics, tools and outcomes: recommends running SonarGraph (free) or SonarQube in CI to measure coupling and structural decay, showing Apache Cassandra with propagation cost 62% and maintainability level 9% as a hard-to-change codebase; explains average component dependency, fan‑in/fan‑out, cyclicity, and cyclomatic complexity thresholds (aim <15); introduces structural depth index 11 vs 550 to compare untangling effort, and shows cycle counts growing from 450 → 900 → 1350 files as a real deterioration signal.
AI-written summary · about
chapters (15)
- 0:00Introduction to metrics
- 1:25Continuous improvement loop
- 3:55SonarQube limitations
- 6:02Cost of poor software quality
- 8:54Microservices pitfalls
- 10:43Why metrics are underused
- 13:58Apache Cassandra metrics
- 16:51Spaghetti code attributes
- 21:24Tools for metric gathering
- 22:48Average component dependency
- 29:08Cyclic dependencies explained
- 36:53Maintainability level metric
- 41:24Measuring code complexity
- 45:19Book and resources
- 45:44Q&A session
related talks
01h 15mHow to measure your architecture
44mTesting Software Architecture
54mResolving technical debts in software architecture
58mTesting Software Architecture
46mMeet the Social Side of Your Architecture
36mHow to make technical decisions?
01h 07mArchitect's Guide to Managing a Code Base
50mBuilding Evolutionary Architectures