We have a database(actually more databases split by primary access path, modulo on a key.) with millions of records (about 400 millions ). Right now I need to save a specific table userData(which has specific data such username, address, city, state, zipcode, country) remotely(on a different server). I'm wondering which would be the best solution given the following requirements :
The data transfer between the host server (database server) and the remote server must be as small as possible.
The operation should not be CPU / resource intensive. It should be silent (should not move needle too much on the monitoring dashboard) due the fact we are using a shared On Demand(as a service) service provider which seems to place a "hold" on our instance when we make too much noise.
The operation must be fast.
I was thinking if it would be possible to do some "compression" on the fly (e.g. using 7z) and I'm wondering what kind of query should I use to make it "silent " to do not disturb the network / CPU too much . 11g database is used on Solaris . On the remote server I'm planning to use redhat but that can be different of course .