On 2017-04-27 09:31:34 -0700, Joshua D. Drake wrote:
> On 04/27/2017 08:59 AM, Andres Freund wrote:
>
> >
> > Ok, based on the, few, answers I've got so far, my experience is indeed
> > skewed. A number of the PG users I interacted with over the last couple
> > years had WAL write ranges somewhere in the range of 500MB/s to 2.2GB/s
> > (max I'veseen). At that point WAL insertion became a major bottleneck,
> > even if storage was more than fast enough to keep up. To address these
> > we'd need some changes, but the feedback so far suggest that it's not
> > yet a widespread issue...
>
> I would agree it isn't yet a widespread issue.
I'm not yet sure about that actually. I suspect a large percentage of
people with such workloads aren't lingering lots on the lists.
> The only people that are likely going to see this are going to be on bare
> metal. We should definitely plan on that issue for say 11.
"plan on that issue" - heh. We're talking about major engineering
projects here ;)
> I do have a question though, where you have seen this issue is it with
> synchronous_commit on or off?
Both. Whether that matters or not really depends on the workload. If you
have bulk writes, it doesn't really matter much.
- Andres