Hi all,
we're working now with ODI and we're moving data to 3 targets: Oracle db, MS SQL Server db and json (Kafka). Our mappings are pretty simple, mainly 1:1. We'd like to have data flows a bit more real time and we are wondering could GoldenGate be our next step. I understood from documentation that for Kafka we need GG BigData. How does it differ from basic GG, is it just an additional licence thing or is there some kind of technical difference? If we buy BigData, does it also work with Oracle and MS?
Another thing that I didn't quite catch, is the GoldenGate CDC concept. Obviously for example for our Oracle line we need to create CDC tables first. I also understood that GG somehow uses db redo logs to catch the changes. If this is the case, then
- why are those tables needed? Does GG save the data from redo logs to the tables by itself or how do we populate the changes there? If GG is only reading the tables, there has to be a way to get something to read there
- how GG knows which data it has already handled? Does it remove the handled data from the tables or what happens to it? What happens if the replication has errors?
Thanks in advance!