15 shopping days left at ALFKI

re: 15 shopping days left at ALFKI

Denis Gobo — Mon, 10 Dec 2007 20:25:35 GMT

Why not 3 DBs?

SmallSample DB

MediumSample DB

LargeSample DB

or is that overkill?

re: 15 shopping days left at ALFKI

Adam Machanic — Mon, 10 Dec 2007 20:28:14 GMT

Which US Census download are you referring to? TIGER?

re: 15 shopping days left at ALFKI

Paul Nielsen — Mon, 10 Dec 2007 20:28:59 GMT

Hi Denis, do you mean small, medium, large in the number of table and complexity or the amount of data?

-Paul

re: 15 shopping days left at ALFKI

Paul Nielsen — Mon, 10 Dec 2007 20:30:43 GMT

Hi Adam,

I found the 1990 Census data easy to download and use, so that's what I'm playing with, but if you have a better idea, I'm open.

-Paul

re: 15 shopping days left at ALFKI

Adam Machanic — Mon, 10 Dec 2007 20:39:08 GMT

Paul,

I'm not familiar with the Census data itself, and didn't even know you could download it. There is a product called TIGER that you can download which contains line segments that describe roads, large buildings, bodies of water, and other formations. I've worked with this data and it is a tremendous pain so I was surprised that you might be considering it :-)

re: 15 shopping days left at ALFKI

Denis Gobo — Mon, 10 Dec 2007 20:40:46 GMT

Paul, I would say size, you can always use a handful of tables. You also don't need to populate all the tables with the same number of rows.....remember not everyone has a TB on their laptop

re: 15 shopping days left at ALFKI

Paul Nielsen — Mon, 10 Dec 2007 20:56:22 GMT

Denis,

I agree, that's why the sample database has to have a fixed loadable set of test data for unit testing and demo code that always generates the same answer to a query.

And a Populate script can pump data into the API so the reader/developer can push the database to whatever size they want for volumn testing, or run multiple copies of the script to test transaction locking and blocking.

-Paul