It's a common question, especially from those at management level in an organization. It's based on the idea that:

"All relational databases have tables, columns and rows ... so we should be able to just chop and change"

That statement (in my view) is basically delusional, and this isn't me being biased for Oracle. Whether you are migrating *to* Oracle, or *away* from Oracle, or for that matter to/away *any* database platform, there are significant differences in the technology.

In regard to 40TB, its not the size of the *database* that matters (because most database platforms can accommodate such sizes), it the size and complexity of:

- the database design (ie, what specific vendor-specific extensions are you using)
- the application code (both inside and outside the database)
- what vendor features are you using without even knowing it (eg, some database lock data when you read it , others don't...that's a HUGE difference)

You just need to look at the complexity and code bases for those large companies that do platform-independent solutions to see how much effort is involved in handling different database back ends.

That money may be better spent investigating options for staying with the Oracle landscape but looking at other opportunities to reduce costs, eg

- changes to hardware
- code optimization
- cloud migration

etc etc

It is smart for *any* company, on *whatever* technology they use, to be seeking ways to get the biggest bang for their buck on their IT investment.