Replication - not all columns

We are needing to setup a replicated copy of one of our Oracle 10g databases to be used as a research/reporting database. One big limitation is that the research database will not have all the columns that the source database contains. For example, one of the tables that is to be replicated can not have the employee socail security # (ssn) column replicated. Since I may not have control over the research databsae, I don't even want to send the ssn to the replicated database, through archive logs or streaming. Keep in mind, some of the tables contain millions of rows.

I am needing some advice on the best method for setting up this type of configuration.

Database size?
Do you need up-to-the-minute replication?
Can you survive with one snapshot a day?
Are your two environments in the same network?
How much bandwidth do you have in between Production and Reporting?

Size does not matter.
Once a day.
Yes.
Bandwidth is not a problem.

I am researching using the Distributed Materialized views. It is a little confusing, specifically on which database will own the MV log and how Owner, Refresher, rights in general are setup.

Quote:

---

Originally Posted by molonede

Size does not matter.

---

really? believe me, when talking about replication size does matter.

Quote:

---

Originally Posted by molonede

Once a day.
It is a little confusing, specifically on which database will own the MV log and how Owner, Refresher, rights in general are setup.

---

If you are going to refresh it once a day MV Logs are inmaterial, you just recreate the MV from scratch in a daily basis.

Size does not matter, what am I supposed to do - tell the users they can only have every other row?

Quote:

---

Originally Posted by molonede

Size does not matter, what am I supposed to do - tell the users they can only have every other row?

---

You don't get it, huh? now it's official, I'm giving up on you.

No I don't get it, that is why I am asking questions. My questions weren't about size of the database, they were about a method in which to replicate a database without replicating specific columns.

Size doesn't matter at all, it doesn't matter to me if its a 10gig database or 100gig database. We can purchase, disk space, memory or processors to satisfy that need. What I need it the best method for replicating the database to a research database that does not have specific columns replicated.

If you want to keep pushing size, then please ignore my questions and let someone that understand that I am asking for a 'method' respond.

Quote:

---

Originally Posted by molonede

...I am asking for a 'method' ...

---

Async Replication method depends on ...
a) frequency,
b) network and
c) size.
I'll let somebody else deal with you.
Have a nice and productive life :rolleyes: