C
ChUIMonster
Guest
If you just use the data server you could break up your application table by table but wouldn't be able to pick and chose what to replicate down to the field level. I think though that the use-case for Pro2 is more around updating a data warehouse than it is for maintaining a transactional DB that is your OLTP system. Pro2 is almost certainly going to have too much latency to be used that way and you wouldn't be very happy if the replication threads stall for some reason. If you have an application with significant OLTP SQL usage (as opposed to mostly reporting) then I would think you would strongly favor the "pure data server" approach. If you are exporting data to a data warehouse sort of system and need to transform it or filter it then I would think that CDC could be your best choice because writing the transformation or filtering logic should be a good fit for 4gl code. I believe that there is some tooling in Pro2 to help with the schema stuff but I'm not very well versed in what the overall schema maintenance burden of Pro2 is. So I cannot really compare them for you.
Continue reading...
Continue reading...