Data and project databases

You have access to two databases. The first is the FellowsSample database, which contains the patient data. The second is a database unique to your project. The project database name starts with an “S'' followed by your project number, which you will find in column A .

The FellowsSample database contains tables with the patient data; however, you do not have direct access to the tables and they will not be visible to you. Instead, a set of project-specific database views have been created in the FellowsSample database. Your project’s views give you read-only access to the subset of data allowed for your project. You can see your views in Object Explorer by opening the FellowsSample database, then the Views folder.

In SQL Server, tables and views are prefixed with a schema name. In the FellowsSample database, the “common” schema has two views named “CONCEPT_DIMENSION” and “I2B2”. These contain metadata about the medical concept codes used in the clinical data. The “S00” schema (replace the “00” with your project number) has views with data specific to your project.

Your project database (S00, where “00” is your project number) is initially empty. Because you cannot write to the FellowsSample database, all temporary tables or aggregate results for your project must be saved to your project database. You can store up to 50GB in your project database. This is not a lot! The full FellowsSample database has over 1TB of data. If you copy data from the FellowsSample to your project database, you will quickly run out of space.

Note that you can use SQL Server temp tables (with a “#” prefix on the table name) in your queries, and that will not use space in your project database. However, temp tables are automatically dropped when your database connection ends.