Archiving

[Laurel]

Hi Erman!
When you have vacation ? :)  Hope you have :)

Do you ha ve experience with Informatica / former Application ?
We doing purge every year using build in utilities. Howether we end-up created clone/archive instance .
So in the long run we will have couple of those... and users dont like it :) as sometimes they need still to run some reports and do that from multiple ones.
Would be much better  if  would be 1  arcive and incremental load ... Too much to ask from oracle, so i am looking at informatica...

Partitioning is another option that is separate pay and there need to do  full investigation how partition .. As affraid design of queries should support it and if we will meet perf issues -> oracle will not support us..
Have you investigated it that ?  Do you have a list of tables and partitioning keys :) ?
I have partitioned cst tables, some queries better , some not.. So I guess I will still purge :)

How are your customers do that ?
Would appreciate your opinion :)
Thanks,Laurel

[ErmanArslansOracleBlog]

Hi Laurel,

My team is administrating a Oracle Financial Analytics Instance, in which the Informatica comes built-in.
So we are dealing with informatica, but we mostly fix the errors in the built-in ETL processes.
And, I have no idea about former application. That's why I can't give you a table level recommendation for that.

I may recommed you Exadata and HCC , but it is also seperate pay:)

archiving seems okay for now.  For querying multiple databases in one go, you can take a look to PDB/CDB functionalities. You can query the same table in multiple different pluggable databases at the same time. However, you need to modify your reports queries a little bit.

Partitioning also fits your case. Once you partition, it will transparent. Pruning will help the performance the queries and the archived data will not be queried when it is not necessary. If you use partitioning, you will not have to host clone instances as well.