> >I'm guessing the high bursts are checkpoints. Can you check your log
> > >files for pg and see if you are getting warnings about checkpoint
> > >frequency? You can get some mileage here by increasing wal files.
> >
> > Nope, nothing in the log. I have set:
> > wal_buffers=128
> > checkpoint_segments=128
> > checkpoint_timeout=3000
> > which I thought was rather generous. Perhaps I should set it even
> > higher for the loads?
>
> But depending on your shared_buffer and bgwriter settings (as well as
> how much WAL traffic you're generating, you could still end up with big
> slugs of work to be done when checkpoints happen.
>
> If you set checkpoint_warning to 3001, you'll see exactly when
> checkpoints are happening, so you can determine if that's an issue.
I did that, and I get two log messages, seeing checkpoints happening
at 316 and 147 seconds apart on my load of a 1.9 GB file. Is this "an
issue"?
Thanks!