Where is Pentaho Kettle's architecture?

Question

Where can I find Pentaho Kettle architecture? I'm looking for a short wiki, design document, blog post, anything to give a good overview on how things work. This question is not meant for specific "how to" starting guides but rather a good view at the technology and architecture.

Specific questions I have are:

1. How does data flow between steps? It would seem everything is in memory - am I right about this?
2. Is the above true about different transformations as well?
3. How are the Collect steps implemented?
4. Any specific performence guidelines to using it?
5. Is the ftp task reliable and performant?
6. Any other "Dos and Don'ts" ?

Answer 1

The Wiki gives a comprehensive list of "how-to" articles, ranging from the basics to specific issues.

Answer 2

See this PDF.