The central tool repository contains the descriptive elements of the archives (relational DB2/Oracle/ SQLServer – File: text fixed records, CSV, VSAM, Sequential, …), for process definition elements that act on them, in a highly detailed coded format, which is usable in the various phases of any subsetting/masking project:
- The administrative phase, which allows the user to “tailor” Daisy, utilizing the repository, according to a principle of minimization of manual intervention. The numerous “wizards” available allow the user to automate repetitive tasks such as discovery of the definition of archives and their relationships or the generation of large numbers of replication processes or extraction. At this stage they also define the masking rules.
- The creation phase, which provides for the customization and automatic generation of batch scripts oriented to the execution platform (operating system and DBMS).
- The implementation phase, in which the generated batch are carried out by the “process engines” specific for the platform, according to the predetermined intervals or on user request.
Daisy provides several strategies for reducing data in respect to completeness and consistency required objectives:
- Selection based on generalized filters made of metadata using specific values or lists of values;
- Selection based on “Driver Tables”, and led by “keys” extracted from higher level subsetting processes;
- Selection based on user-defined complex Administrator filters;
- Total replication (for particular classes of archives, eg.: domain tables).