Re: Load Distributed Checkpoints, take 3

Heikki Linnakangas <heikki(at)enterprisedb(dot)com> writes:
> Tom Lane wrote:
>> Who's "we"? AFAICS, CVS HEAD will treat a large copy the same as any
>> other large heapscan.

> Umm, I'm talking about populating a table with COPY *FROM*. That's not a
> heap scan at all.

No wonder we're failing to communicate. I assumed you were talking
about copy-to-file. Copy-from-file is going to be creating WAL entries
hence the no-checkpoint case doesn't apply anyway, no?

[ thinks ... ] Oh, you must be positing the case where the recently
added skip-WAL-if-table-is-new-in-this-transaction optimization applies.
Well, that thing could probably do with some more work anyway (I wonder
why it's using shared buffers at all anymore). I don't really think
that case should be allowed to drive our thinking about how the bgwriter
should work.

regards, tom lane