| From: | Peter Geoghegan <pg(at)bowt(dot)ie> |
|---|---|
| To: | Patrik Novotny <panovotn(at)redhat(dot)com> |
| Cc: | PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: RFE: Make statistics robust for unplanned events |
| Date: | 2021-04-20 21:25:57 |
| Message-ID: | CAH2-WzkxhUBpVyQompg=iG1=SQpMxRiZ7f0AyJupOmdoojeUtA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Tue, Apr 20, 2021 at 5:00 AM Patrik Novotny <panovotn(at)redhat(dot)com> wrote:
> As far as I've checked, this would have to be implemented.
>
> My question would be whether there is something that would make this impossible to implement, and if there isn't, I'd like this to be considered a feature request.
I agree with you.
Maybe crash safety would require some care in cases where autovacuum
runs very frequently, so that the overhead isn't too high. But
overall, non-crash-safe information that drives autovacuum is penny
wise, pound foolish.
I'm sure that it doesn't matter that much most of the time, but there
are probably workloads and use cases where it causes significant and
persistent problems. That's not the right trade-off IMV.
--
Peter Geoghegan
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Stefan Keller | 2021-04-20 21:29:43 | Re: ML-based indexing ("The Case for Learned Index Structures", a paper from Google) |
| Previous Message | SATYANARAYANA NARLAPURAM | 2021-04-20 21:19:48 | Re: Synchronous commit behavior during network outage |