| From: | Andres Freund <andres(at)anarazel(dot)de> |
|---|---|
| To: | "tsunakawa(dot)takay(at)fujitsu(dot)com" <tsunakawa(dot)takay(at)fujitsu(dot)com> |
| Cc: | "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Make ringbuffer threshold and ringbuffer sizes configurable? |
| Date: | 2020-02-06 05:19:09 |
| Message-ID: | 20200206051909.dmyvh4ivz2l4nvyl@alap3.anarazel.de |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi,
On 2020-02-06 05:12:11 +0000, tsunakawa(dot)takay(at)fujitsu(dot)com wrote:
> I think per-table reloption is necessary as well as or instead of GUC, because the need for caching depends on the table (see below for Oracle's manual.)
I'm inclined to not do that initially. It's going to be controversial
enough to add the GUCs.
> I'm afraid it would be confusing for a user-settable parameter to have
> different units (percent and size). I think just the positive
> percentage would suffice.
IDK, I feel like there's good reasons to use either. But I'd gladly take
just percent if that's the general concensus, rather than not getting
the improvement at all.
Greetings,
Andres Freund
| From | Date | Subject | |
|---|---|---|---|
| Next Message | keisuke kuroda | 2020-02-06 05:25:03 | In PG12, query with float calculations is slower than PG11 |
| Previous Message | Michael Paquier | 2020-02-06 05:15:02 | Re: typos in comments and user docs |