Re: Multiple Tables Split By Date
Hi Rick,
Agreed it does take its toll using UUIDs. We do store them as Binary(16) in the DB which helps.
We can never be sure to write to one MASTER at a time. This is because our application runs from three locations for redundancy. In a situation where one location cannot talk to the other, we have a problem whereby there are writes to different masters which causes the issue with auto_increment IDs.
This is why i think we will see a good performance advantage with partitioning as we are not loading up so much of the data at once. Most of the queries are date based, and are searching a few months back on average.
Once we have run some performance benchmarks we'll know for sure!
thanks again
Subject
Written By
Posted
February 20, 2013 05:40AM
February 21, 2013 11:23PM
February 22, 2013 06:30AM
February 23, 2013 12:10PM
February 25, 2013 03:07AM
February 25, 2013 11:48PM
Re: Multiple Tables Split By Date
February 26, 2013 02:26AM
Sorry, you can't reply to this topic. It has been closed.
Content reproduced on this site is the property of the respective copyright holders.
It is not reviewed in advance by Oracle and does not necessarily represent the opinion
of Oracle or any other party.