On Wed, 13 Oct 2010, Tom Lane wrote:
> Neil Whelchel <neil.whelchel@gmail.com> writes:
>
> That's probably got little to do with caching and everything to do with
> setting hint bits on the first SELECT pass.
>
> I concur with Mark's question about whether your UPDATE pushed the table
> size across the limit of what would fit in RAM.
Neil, can you just double the size of your initial test to make sure that
it's too large to fit in ram to start with?
David Lang
