[wdmmg-dev] Online Analytical Processing (OLAP) for WDMMG

[Stefan Urbanek]

Hi,

I have put together couple of notes about online analytical processing (see [1] and [2]) for WDMMG:

Main point is to introduce metadata: create a logical model that reflects data *how they are being measured*, that is: from user's perspective. 

Documents with notes are here:

Online Analytical Processing and Aggregations (document for specifying aggregation design and API):
	http://bit.ly/gwQTUW

Spreadsheet with WDMMG metadata (list of fields in source DB, proposed multidimensional model):
	http://bit.ly/gUNbuq

Preliminary proposal (diagram):
	http://bit.ly/ikj8Yq

I started porting an OLAP framework from ruby to python [3]. The logical model in python version is close to the Oracle 11 OLAP [4] way of modelling. Currently the port handles logical model [5] and creation of denormalized view (in relational database) based on the model. I hope to have aggregation browser ready soon. Also I will provide more detailed documentation with examples.

I would like to know, what do you think about it?

If you have any questions or want more explanations or examples, just let me know.

[1] http://en.wikipedia.org/wiki/Olap
[2] http://en.wikipedia.org/wiki/OLAP_cube
[3] https://github.com/Stiivi/brewery-py
[4] http://bit.ly/fw4wdR
[5] http://databrewery.org/doc/cubes.html

Regards,

Stefan Urbanek