"create table as" statement versus an "insert into" statement (performance

I could use some help.

Does someone know why a create table as statement would perform better than a truncate table, and then insert into statement? Gut instinct tells me the create table is cleaner and the insert into has to find free chunks, but that can't be all of it.

The background is that we have some developers who are executing large data loads and have to recreate objects every evening as part of a hierarchical OLAP tool process. They are looking to shave time wherever they can to get it all into their batch window.

I am not sure where to begin to look for information that would show why one way would perform better than the other. I think it would be under parsing algorithms, which would be cool to learn if there was a list of them or a chart.

Thanks in advance, Received on Mon Oct 23 2000 - 16:52:30 CDT