I'm looking to extract about 40 Million rows out of a DB2 database. My question has to do with which method will yield the best performance. I can use JDBC to run a select statement via Java. Or I can use the extract command to extract the data in XML. Which choice will be the fastest? Are there any other options available?

What do you want to do with the data once it is "extracted"?

Oh, and it may be a good idea to tell us which DB2 version you are using on which platform.

The data is being passed to a 3rd party program. This program has facilities to bring in data via JDBC, but I've found that the extracts take about 6 hours. The other option is to extract via XML, but I'm not sure how fast that will be.

I'm using db2 8.1 on aix