4.12 Database Conversion Plan (DCP)

---

The DCP is used to define existing application files that need to be converted, specify the new files that need to be created, identify the volume of data that is to be converted, and stipulate which organizational entities are to be involved in defining the conversion process, target dates, data integrity, and so on. This document is an end-to-end flow for data placed in and retrieved from the system. In most instances, existing data from legacy systems will be integrated with new systems data and data structures. To successfully accomplish this task, much data analysis needs to take place. A Data Analysis Team is required to determine the need to convert data from one internal machine representation to another.

Part of the responsibility of a Data Analysis Team is to evaluate all of the various data elements originating from several files and databases and validate the content of the fields. It is often necessary for data analysts to merge different fields from different files or databases into a single data file. Part of the data analyst's job is to establish relationships or dependencies among different data fields currently stored in different files and/or databases and to synchronize several files and databases concurrently. He or she must decide to expand or shorten the size of data elements and reorder the records or place data elements in a different sequence to accommodate new data structures. Sometimes, data analysts must modify the contents of the data field itself.

Data analysts are required to analyze the high-level data requirements of the new system and identify the primary domains of change that exist between the current and new system data architectures. They must examine the overall quality and integrity of the data presently stored in manual and automated files and databases belonging to the current system. The analysts must identify existing or special software tools and facilities or hardware and networking equipment that will be required to support the data conversion process.

After all of this work is completed, the Data Analysis Team will derive an optimal strategy to perform a well-defined data conversion effort in the context of the project scope. The magnitude of this conversion effort is directly related to the complexity of the existing and new system data architectures, the volume of data involved in the conversion process, and the degree of change associated with the application to be installed. The Data Analysis Team must walk through the preliminary system data conversion strategy. These analysts are required to evaluate the scope of change between the old and new systems and to identify all existing application files that need to be converted, along with a brief description of the records and data elements involved in the process. The new files that must be created from scratch need to be identified, along with a brief description of the data elements that must be loaded into these files. For each file, the estimated volume of data that needs to be converted to accommodate the new system data requirements should be documented.

During the data conversion process, systems developers may be called on to provide unique data conversion programs, data validation programs, or file comparison facilities that assist in converting the existing files and databases efficiently. They may be asked to produce reports needed by the users to validate the results. For these reasons, it is very important that the major objectives and scope of the conversion process, along with the major issues to be addressed by the development team and the users, be explained at the beginning of a data conversion process. The roles and responsibilities assigned to each user and systems group involved in the conversion process should be identified.

The description of the files and databases that must be converted in association with this particular release of the system should be documented and distributed to all interested parties. This documentation should provide a section that details the type of conversion that will be performed against each affected file or database:

• Automated-to-manual conversion (not desirable but often necessary)

• Automated-to-automated conversion

• Fallback procedures in case something goes wrong during the conversion process

• Manual-to-automated conversion

• Manual-to-manual conversion (the worst way to go)

• The type of reports required by users to verify the results of the conversion

• The type of reports required by users to analyze the quality of the data

• The particular sequences in which the files will be converted

• The dependencies that exist among the files

The major business units of the organization that should be involved in the conversion process are, at this point, required to identify possible constraints that might be imposed on the conversion process. The Data Analysis Team must provide them with expected target completion dates. They should stipulate any special handling of highly sensitive data and identify any dependencies with other systems and, if needed, state the necessity to construct interfaces to them.

During the data conversion process, the Data Analysis Team most often needs to ensure availability of the files to the users (i.e., the current production files must remain available for the users 23 hours a day, 5 days a week) and to create an availability schedule for public dissemination. During the conversion process, representatives from the business owners should periodically examine the level of quality of the data currently stored in the existing files. Ultimate responsibility for ensuring the success of the conversion and the integrity of the data lies with the users.

---