School of Information Systems

Working With in-Database Data

Database processing enables blending and analysis against large sets of data without moving the data out of a database, which can provide significant performance improvements over traditional analysis methods that require data to be moved to a separate environment for processing. Performing analysis in the database can save processing time.

With TIBCO Spotfire, if a column is added to a database table you need to perform a Refresh Schema operation in Spotfire in order to see the new column in the analysis. Not all users will have the sufficient database privileges to perform a full schema refresh. However, changes to the actual data can be updated using a simple Reload/Refresh Data by most users. But, there are some limitations on using TIBCO Spotfire:



The table visualization and the Details-on-Demand (D-o-D) cannot display more than 10000 rows from an in-database data source. The table and the D-o-D do not support data from cubes. Virtual columns are not supported for in-database data. There is a preset limit of 10000 rows to prevent large databases from overflowing Spotfire.
Marking in table visualizations is only supported if the underlying database table has a primary key defined. A primary key is required to know which rows to mark.
The summary table cannot be used with in-database data. The summary table has not yet been adjusted to be able to handle in- database data.
Not all of the standard Spotfire aggregation methods and expressions are available on all external systems. On the other hand, there may be other methods available that Spotfire does not have by default. The aggregation methods supported by the external data source determines which methods will be available for in- database data.
Automatically created date and time hierarchies are not available for in- database data. The shortcuts for setting up date and time hierarchies have not been adjusted to handle in- database data.
If the connection to the external data source is lost, no further analysis using that data source can be performed. Since no external data is stored within the analysis in runtime and each change induces a new query to be sent to the external data source, the data source must be available as long as changes to the visualizations are to take place.
You cannot delete columns or rows from in-database data tables. In- database data tables contain no data in runtime so nothing can be deleted.


Tri Nur Auliyaa