Data security
A Business Intelligence solution has two conflicting aims:
-
To make the information as accessible and easy to understand as possible
-
To ensure that this often confidential and key information is only accessible
to the relevant users.
Often, enthusiasm to satisfy the first aim means that the need to build a
strong, robust and manageable data security mechanism falls between the cracks.
In addition:
-
data security can be complex e.g. only let the Sales Manager for Christchurch
look at sales figures in the Christchurch region but don‘t let him\her see the
distribution costs
-
security can be implemented in multiple portions of the application
-
Database
-
Cube
-
Reporting/query product levels
-
security can be implemented very differently between two similar products e.g.
two database products.
The common assumption, that data security can be implemented at the repository
level as a post build activity, often turns out to be naïve
when no up front analysis has been performed.
Read next »
