Following up with the follow up of Trivadis Tech Event :-).
In addition to the five minute talk about Linux Tracing for Oracle, I presented on a quite different subject: “Object Relational Mapping Tools – let’s talk to each other”.
My original intention was to write 2 – 3 successive posts on the subject, but as always, time’s scarce and so I’ve decided to just upload the presentation, see here.
But allow me some words on the background. A database admin (though spare time coder, too ;-)), presenting on O/R Mappers? How could that be? At least then, these performance killers should really get pulled to pieces, right?
Well, no. They aren’t. Instead, my intentions have been the following:
- Explain to a DBA audience why developers might want to use them.
- Explain – to DBAs and developers both – what are probable difficulties, on a conceptual level (the infamous “Object-Relational Impedance Mismatch”)
- Explain – again on a conceptual level, but with concrete examples, and to both audiences, too – what to watch out for when fetching data (i.e., doing SELECTs as opposed to data manipulation).
“What to watch out for” – this has actually been the talk’s main purpose. (I plan to follow up with load tests / “real empirical stuff” at a later time).
And how about the “let’s talk to each other”? You may have guessed already …