- Fast – Hundreds of MB r/w per second from thousands of clients
- Central data backbone – Data streams are partitioned and spread over a cluster of machines
- Durable – Messages are persisted on disk and replicated within the cluster
- Fault-tolerance – Cluster-centric design
Writing code that uses external APIs is not trivial and testing it neither. This post describes how to drive code around external APIs initially and is the second part of How to drive code by sketching it. The test-approach is only one of a couple of possibilities and used just to outline how to drive code via Quick-Fix of an IDE like eclipse. Weiterlesen
Actually, there are numerous of ways to sketch software or pieces of it. Drawing with a pen on a paper or on a white-board, using an UML tool et cetera. The goal is just to get things simplified and get an abstract view of it to find the core issues. With the introduction of Test-driven Development (TDD) a new way appeared end of the 90ties. TDD is controversially discussed and I don’t want to dig into the pros and cons of this approach. Rather I’d like to pick up one aspect of this methodology: drive code and design by coding.