More on the 9.2 to 10.2 upgrade

Sunday, September 9th, 2007 by Peter Scott

But first. It’s good to see that two bloggers in the Washington DC area are back on-line. Welcome back Tom and Sponge
A few posts ago I mentioned my final 9 to 10g upgrade (unless someone else wants me to do a 9 to 10 upgrade project). The mechanics of the upgrade are done, I […]

Upgrading to 10g - OK, I’m a bit behind

Friday, August 31st, 2007 by Peter Scott

At long last, my final data warehouse customer on Oracle 9 is moving on to 10g. It has been a somewhat protracted move as for reasons of vendor support certification we had to adopt a specific order of refreshing their whole technology stack. That’s not to say we have no Oracle 8i customers out there […]

Reading time - Oracle 11g

Wednesday, August 15th, 2007 by Peter Scott

There you are - I’ve mentioned it. But I have not downloaded it yet.
One of the things I do with any new release is to read some (no, not all) of the documentation. The new features documentation, of course, is a given and enough of the of the other stuff to have a good idea […]

Moving partition statistics

Wednesday, August 8th, 2007 by Peter Scott

Over in another place, Doug Burns has been writing on the joys of reproducing a full size database as test instance. He has even (graciously) put up with my wittering about things that not relevant to his articles. So as not to clutter his blog with a stream of off-(his)-topic posts here is a […]

Oracle Partitioning 2007

Saturday, August 4th, 2007 by Peter Scott

I see that Oracle have posted a series of white papers on the soon to arrive 11g release, the one on partitioning caught my eye.
But first an aside: Does anyone actually use slower storage for less-accessed data? The theory sounds good: put the-once-in-a-blue-moon stuff on slower, cheaper, disk and the “use all of the time” […]

Playing Around with Star Transformations and Bitmap Indexes

Friday, July 27th, 2007 by Mark Rittman

I’ve been doing some work this week with a client in London, who wants to put together a reporting data mart based on Oracle Database 10gR1. Although they and I were fairly sure a dimensional database design using bitmap indexes and star transformations would give them the query performance they required, there were a […]

Look Before You Leap

Saturday, July 21st, 2007 by Mark Rittman

A recent post on Howard Rogers’ blog around the hype and enthusiasm around the launch of Oracle 11g, together with some feedback I got during the review process for my OU seminar slides, got me thinking about how we sometimes rush in to adopt just-released versions of software, and how we can sometimes over-emphasise new […]

Campaign against incorrect data types

Monday, July 16th, 2007 by Peter Scott

Those that have met me know that beneath that benign, placid exterior lurks a data-geek of great passion. Show me the work of others where they reinvent the wheel by writing their own procedural methods (in PL/SQL or T-SQL or whatever) to replicate a native function of the database or store numbers as characters, or […]

A Day of Product Releases

Wednesday, July 11th, 2007 by Mark Rittman

Today was the official launch of Oracle Database 11g in New York City, whilst another event, the Oracle Fusion IT Conference, was being held at the Emirates Stadium in London. I couldn’t make it to either as I was onsite with a customer, but the New York event marked the public launch of Oracle’s new […]

I know that tune

Tuesday, July 10th, 2007 by Peter Scott

Certain solitary acts give me great pleasure. Oddly, I consider the intellectual challenge of getting some of my customers’ queries to fly a solitary act, it’s me against the database. Or perhaps me against the customer.
I have been working on a set of poorly performing queries that have been causing a customer a lot of […]