Prologika Forums

Hi Teo,

1. I think that in your book: "applied analyses services.." figure 9.3 does not correspond to the text at the bottom of the page. Shaded cell is not the one described by the coordinates.

2. I have 6 client databases, havingthe the same schema. I have a separate database that defines hierarchy for each client (one table). Most of the reports will not cross boundaties of the inhdividual database, but "nice to have" capability of cross comparison reports. I would like to have one cube instead ot 6 (mostly from the management point of view). Will partitioning work in this scenario. I am planning creating one cube with separate partition for each client. Do you see anything wrong going this direchtion ? 

1. You are right of course. Thank you for reporting this. I need to update the book errata.

2. This is something I haven't tried but I can't think of a a reason why this shouldn't work. You primary data source need to point to the database with the common client hierarchy. You need to set up 6 additional data sources pointing to the client databases. Each partition in the cube will be bound to a corresponding "client" fact table. I am just not sure what other conformant dimensions you will need to slice the cube and where their data will come from.