DropBear wrote on May 29
th, 2015 at 11:33pm:
Maybe bro, sorry if I got the terminology wrong.
Without seeing the issue, I'm not sure what we're dealing with in terms of systems or the best way to achieve it.
They have an existing server merge tool that likely needs updating.
The guys here have speculated what the backend might be, so a DBA might be able to do the job with some scripts.
I'm optimistic to the extent that I have seen far more complex data amalgamations done for businesses.
A couple of game servers is unlikely to be more difficult than many other business critical data amalgamations during mergers and takeovers.
Having a plan, the right people and a decent budget I'm sure it can be done.
Another speculation : DB servers iseems to be MS-SQL...
Since I've spent years tinkering with Sybase ASE ( and Replication Server ), I suspect I have a clue or two on how to merge two databases into a single one.
It all depends on the datamodel used... and honestly, I don't think the DDO datamodel is merge friendly... Because I don't think the DDO datamodel was created by a real DBA.
They picked up the guy that had
Knows SQL on his resume and named him DBA. They might have had a real DBA for the GLS server, but this server is global for the company, so he just gave the queries/stored procedures entry points and results to the DDO team.
I wasn't interested in it at that time ( and since I came from Europe, it was already on it's way out when I was dumped in G-Land ) but the API could have given me a clue about the datamodel. now, since the API server is dead, I can't really get the information anymore.
Anyway : TL;DR : the ease of merging two databases together is tied to the datamodel and if the datamodel sucks, merging databases will be a nightmare.