On Tue, 2005-11-01 at 10:37 -0500, Ron Peacetree wrote:
> I'm surprised that no one seems to have yet suggested the following
> simple experiment:
>
> Increase the RAM 4GB -> 8GB, tune for best performance, and
> repeat your 100M row insert experiment.
>
> Does overall insert performance change? Does the performance
> drop <foo> rows in still occur? Does it occur in ~ the same place?
> Etc.
>
> If the effect does seem to be sensitive to the amount of RAM in the
> server, it might be worth redoing the experiment(s) with 2GB and
> 16GB as well...
Ron,
I would like to try this, however, since I'm sitting about 1000 miles
away from the server, tweaking things is not as simple as one might
hope. I would also like to understand what is going on before I start
changing things. If I can't get a satisfactory explanation for what I'm
seeing with current hardware, I'll have memory added and see what
happens.
-K