GouthamAnand
asked on
oracle tables replication
Hi,
I have two tables TABLE1 and TABALE2 which I joined and build a query. Now the data fetched with this query I need to insert or update(if record already existing) into a table TABLE_REP which is in another schema but same database.
First time I need to load all the data after that for every week I need to load the data into the new table TABLE_REP. For a week data may be appx. 10 million records.
One way is we need to write a Stored procedure and schedule for every week after the first full load.
Can you please suggest is there any other better way?
Can we use Materialized view and schedule it for every week after the first complete refresh? In this way can we track and log any failed records in any log table?
Also I need to archive the data from the TABLE_REP table for older data into another table and delete from this table. So if I use Materialized view can I achieve this also?
Like this I have 10 reporting tables for which data needs to be fetched for every table with 2 base tables joining from another schema.
Which option is better out of these two approaches (stored proc and MV)?or Is there any better approach also other than these two options? Request you to suggest.
Thanks,
I have two tables TABLE1 and TABALE2 which I joined and build a query. Now the data fetched with this query I need to insert or update(if record already existing) into a table TABLE_REP which is in another schema but same database.
First time I need to load all the data after that for every week I need to load the data into the new table TABLE_REP. For a week data may be appx. 10 million records.
One way is we need to write a Stored procedure and schedule for every week after the first full load.
Can you please suggest is there any other better way?
Can we use Materialized view and schedule it for every week after the first complete refresh? In this way can we track and log any failed records in any log table?
Also I need to archive the data from the TABLE_REP table for older data into another table and delete from this table. So if I use Materialized view can I achieve this also?
Like this I have 10 reporting tables for which data needs to be fetched for every table with 2 base tables joining from another schema.
Which option is better out of these two approaches (stored proc and MV)?or Is there any better approach also other than these two options? Request you to suggest.
Thanks,
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Thank you.
Even though you already accepted an answer to this I have some comments.
First:
Streams replication is free in Enterprise Edition.
If the tables need to be copies of each other, why not a view in schema2 that selects directly from schema1?
I would look at materialized views.
>>Populating a materialized view adds load to server.
So does physically inserting and updating manually. Same amount of data has to move from pointA to pointB so I'm not sure the manual process is better or worse than a fast refresh MV.
I would experiment with both options and choose the one that is most efficient. We cannot predict which method will work best on your systems.
First:
Streams replication is free in Enterprise Edition.
If the tables need to be copies of each other, why not a view in schema2 that selects directly from schema1?
I would look at materialized views.
>>Populating a materialized view adds load to server.
So does physically inserting and updating manually. Same amount of data has to move from pointA to pointB so I'm not sure the manual process is better or worse than a fast refresh MV.
I would experiment with both options and choose the one that is most efficient. We cannot predict which method will work best on your systems.
ASKER
Thank you slightwv.
you can consider this option too if you cannot afford going for licensed version software..
Unloading and Loading are not database procedural level operations they are done from OS level, so it may be performed on schema level, however as this operation also happens on the same server.
check out the site for more details on loading and unloading the data using oracle query
http://www.oracle-developer.net/display.php?id=325