On the other side of the represented flow, there is a database storing the extracted informations on the repository structure and the audit data organized in a specific Data Mart. By it’s nature, a Data Mart is a structure that is usually oriented to a specific business line or team and, in this case, represents the audited actions in the Alfresco E.C.M. The implemented Data Mart develops a single Star Schema having one only measure (the number of audited actions) and the dimensions listed below:
- Alfresco instances to manage multiple sources of auditing data.
- Alfresco users with a complete name.
- Alfresco contents complete with the repository path.
- Alfresco actions (login, failedLogin, read, addAspect, etc.).
- Date of the action. Groupable in day, month and year.
- Time of the action. Groupable in minute and hour.
For the ones of you that doesn’t feel confident with the Data Warehousing techniques, the developed Star Schema is able to reply to most of the business queries a user could do on the stored data. For that reason we think the used Data Warehousing’s techniques are the right choice to fully analyze the Alfresco audit data (and not only).
From a physical point of view the Audit Data Mart is a DBMS composed by dimension tables and fact tables. The A.A.A.R. Data Mart is stored in the ‘AAAR_DataMart’ DBMS together with the working tables used for the solution and detailed ahead in the documentation.
A.A.A.R. Data Mart is developed on PostgreSQL v9.0 and MySql Community Server v5.5 but should work on different versions of the DBMS (also older) because no particular features are used. Please, suggest limits and tests to let it work even in other contexts.