30 Jan 10 Data Subsets and Databases

Data Subsets are marted into local databases in order to satisfy the majority of the queries, while an additional set of summary tables are replicated into each mart in order to provide a composite view. A number of user access tools, particularly specialist data mining or multidimensional analysis tools, may require their own internal database structures in order to function. In practice, these tools create their own data mart designed to support their specific functionality, but the issues they encounter are common to most of these tools.

For example, it is very difficult in practice to operationally support more than a handful of data marts. You will find that the time it takes to populate the mart as part of the load process can be significant. If you try to populate more than a handful, you will probably run out of time window. This is a common problem when using a number of user access tools that require data marts.
This problem will exist even if the organization is using multiple copies of a single data-marting user access tool. The issue is the lead time to populate the data in the mart; even if the data structures are identical, the source data will be different and will take time to load.

Leave a Comment

You must be logged in to post a comment.