News Stay informed about the latest enterprise technology news and product updates.

Oracle Database 11g makes waves at Burlington Coat Factory

The CTO of Burlington Coat Factory talks about data center failover and the new features of Oracle Database 11g.

Oracle Database 11g is getting some early use at one major New Jersey-based clothing maker and retailer. In part two of our interview with Mike Prince, the CTO of Burlington Coat Factory -- which has two data centers ten kilometers apart -- talks more about the technologies and services that Oracle offers to help automate data center failover, as well as the key features of Database 11g that his company is interested in.

What does Oracle offer that could help you further automate failover?

Mike Prince: Well, the tool you'd use to control it would be [Oracle Enterprise Manager]. But there's just things [and services they offer] to facilitate failover. We have failed over to Data Guard instances, but it's more manual. You can set this thing up to be all scripted. You can have Oracle watchdogs that would detect the need to do it and actually automatically trigger it. But we haven't gone that far yet. Part of it was just a reluctance to give that much to Oracle right off the bat. This sort of stuff was all new in [Oracle Database 10g] and the other thing was we didn't want to get all this stuff worked out in order to get to 10. So we cut corners on the install and it just hasn't been a priority. Besides, we have great reliability [and availability] and we don't intend to lose a whole data center.
More on Oracle Database 11g:
Oracle Database 11g and the information management challenge

Installing Oracle 11g on Linux

What other features of 11g are you interested in?

Prince: But one of the things that we could benefit from was if we could get use out of the standby databases, and there's two ways to do that in 11g. There's a thing called Active Standby, which gives you an up-to-the-minute copy of the production database that you can open in read only mode while it's doing the standby. And then they have a thing called Snapshot Standby. There, you keep shipping the logs across so you get the data protected in the other data center but you don't apply it while you have the database open for use. And going that way is a no-price option.

What are the key benefits of this approach?

Prince: The big benefit [other than getting use out of standby databases] would be that if I'm using they cycles over here, I can justify having the same number of nodes configured for a database in both locations. That way if I do need to switchover, where it might take me awhile to roll forward depending on how far into the standbys thing I was, it would be a lot shorter than it is right now because I've got to basically bring up a lot of instances in [what basically is a] semi-manual process. So [compression capabilities, Active Standby and Snapshot Standby] were the things that excited us when we tested Oracle Database 11g, and those things worked well.

Could you quantify the results of your test of the data compression feature in Database 11g?

Prince: We took one of our problematic tables that was very big and we got 2.5x compression out of the data compression feature. So we predict enormous savings with that. But, basically, you divide whatever size it is by 2.5 and that's what you end up with.

Read Part 1 of the

Dig Deeper on Oracle database design and architecture

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.