Unusual performance question

I am importing several million sales records in a test run using the derby database.  I let it run all night and it averaged about 8000 records an hour and my CPU utilization averaged under 10%.

I stopped and restarted it to clear out the log files and when I restarted it, my CPU utilization is now 95% and I am processing 80000+ records an hour.

I did nothing else except for deleting the contents of the logs directory.

This is running on MS Windows.

Anyone have a clue as to what could explain the performance increase (so I can duplicate it again)?

Skip

[BJ Freeman]

might want to ask in the derby ML

[hidden email] sent the following on 11/9/2007 6:09 PM:

> I am importing several million sales records in a test run using the derby
> database.  I let it run all night and it averaged about 8000 records an hour and
> my CPU utilization averaged under 10%.
>
> I stopped and restarted it to clear out the log files and when I restarted it,
> my CPU utilization is now 95% and I am processing 80000+ records an hour.
>
> I did nothing else except for deleting the contents of the logs directory.
>
> This is running on MS Windows.
>
> Anyone have a clue as to what could explain the performance increase (so I can
> duplicate it again)?
>
> Skip