| From: | Heikki Linnakangas <hlinnaka(at)iki(dot)fi> |
|---|---|
| To: | Peter Geoghegan <pg(at)bowt(dot)ie>, Pavel Luzanov <p(dot)luzanov(at)postgrespro(dot)ru> |
| Cc: | Melanie Plageman <melanieplageman(at)gmail(dot)com>, "pgsql-generallists(dot)postgresql(dot)org" <pgsql-general(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: PG17 optimizations to vacuum |
| Date: | 2024-09-02 21:11:48 |
| Message-ID: | d10adba2-e028-44c1-b5a5-96ff922f546f@iki.fi |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On 03/09/2024 00:01, Peter Geoghegan wrote:
> On Mon, Sep 2, 2024 at 4:58 PM Heikki Linnakangas <hlinnaka(at)iki(dot)fi> wrote:
>> Do you have any non-default settings? "select name,
>> current_setting(name), source from pg_settings where setting <>
>> boot_val;" would show that.
>
> What about page checksums?
>
> One simple explanation is that we're writing extra FPIs to set hint
> bits. But that explanation only works if you assume that page-level
> checksums are in use (or that wal_log_hints is turned on).
Hmm, yeah, that might be it. With chceksums enabled, I see ~120k WAL
records, vs ~90k without checksums. But there's no difference between
v16 and master.
Pavel, did you test v17 with checksums enabled and v16 with checksums
disabled, by any chance?
--
Heikki Linnakangas
Neon (https://neon.tech)
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Peter Geoghegan | 2024-09-02 21:17:30 | Re: PG17 optimizations to vacuum |
| Previous Message | Peter Geoghegan | 2024-09-02 21:01:54 | Re: PG17 optimizations to vacuum |