Data Migration Services Case Study

Customer
Unite provides Student accommodation across the entire country. Currently, they provide homes for around 38,300 students in the UK. Unite use a variety of systems for the administration of rental properties, rent processing, fees and accounts administration.
Customer Situation
Unite hosted a very old Unix server running a Informix property management database that contained many years worth of accounting data. The company that originally supplied the database and servers no longer existed. The hardware had begun to fail, so they needed the database and all data migrated to a new Oracle solution for historical reporting and regulatory compliance.
Solution Provided
We provided a full migration service. All databases, tables, columns and data was migrated from the old Unix system to a new Windows based server architecture running Oracle 10g. As part of the solution, we provided full documentation of the databases, tables and contents migrated, as well as a Cognos Reporting system for bespoke reporting from the new database. As part of the migration, we amalgamated 128 different databases into a single database, allowing across-company reporting.
Benefits of Service
As a result of the migration, the old Unix server was decommissioned. This resulted in the removal of hosting costs, support costs and hardware costs. At the same time, the historical information was rescued and placed into a format which Unite can continue to administrate.
Methods Utilised
As the suppliers of the original systems (Informix and the administration application) no longer existed, we utilised database ‘sniffing’ technology which identified the databases present on the server, tables within the databases and their structures. Data Migration routines were then created using Extraction, Translation and Loading (ETL) tools and the data migrated in two runs (1 test, 1 live). As part of the extraction, we created checksums of data out and data in, and table, row and byte count levels to ensure all data was migrated.
External Factors
None

Data Transformation
Data Cleansing
Multiple Sources
