Data migration projects suffer from the same issues as “waterfall” software deliveries: they isolate and serialise analysis and profiling, mapping specification and implementation, data load and target system testing.
Since the true measure of migration success is whether the target system functions correctly and because in reality the legacy systems continue to be in use throughout the project’s lifetime, this approach maximises the probability of late risk materialisation and therefore project overrun or outright failure.
Agile data migration is difficult because of close coupling between the source and target system data models. This close coupling makes it hard - if not impossible - to make small incremental changes to data mappings and transformations in the context of a short-cycle agile process that is responsive to the continuous change that is inevitably present in all but the smallest enterprise environments.
Ontology Integrity Manager uses semantic search in the place of traditional data integration and transformation approaches to reduce this coupling and places a flexible model between the source and target that can be changed and tested quickly and inexpensively.
This is turn makes the application of agile methods to data migration a natural and cost effective way to ensure project success.
The modelling technology used by Ontology does not rely on relational database schemas and therefore requires no big up-front model design. It directly supports iterative, incremental solution implementation, enabling fast, value-first, cost effective solution delivery with minimal impact on your organisation.