| From: | Heikki Linnakangas <hlinnaka(at)iki(dot)fi> |
|---|---|
| To: | Michael Paquier <michael(at)paquier(dot)xyz> |
| Cc: | Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>, Jeff Davis <pgsql(at)j-davis(dot)com>, vignesh C <vignesh21(at)gmail(dot)com>, Peter Smith <smithpb2250(at)gmail(dot)com>, Tomas Vondra <tomas(dot)vondra(at)enterprisedb(dot)com>, "Hayato Kuroda (Fujitsu)" <kuroda(dot)hayato(at)fujitsu(dot)com>, Shubham Khanna <khannashubham1197(at)gmail(dot)com>, Dilip Kumar <dilipbalaut(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> |
| Subject: | Re: Improve eviction algorithm in ReorderBuffer |
| Date: | 2024-04-10 23:03:08 |
| Message-ID: | a426481b-8622-4bb4-a7e5-d236336a6662@iki.fi |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 11/04/2024 01:37, Michael Paquier wrote:
> On Thu, Apr 11, 2024 at 12:20:55AM +0300, Heikki Linnakangas wrote:
>> To move this forward, here's a patch to switch to a pairing heap. In my very
>> quick testing, with the performance test cases posted earlier in this thread
>> [1] [2], I'm seeing no meaningful performance difference between this and
>> what's in master currently.
>
> Reading through the patch, that's a nice cleanup. It cuts quite some
> code.
>
> +++ b/src/include/replication/reorderbuffer.h
> @@ -12,6 +12,7 @@
> #include "access/htup_details.h"
> #include "lib/binaryheap.h"
> #include "lib/ilist.h"
> +#include "lib/pairingheap.h"
>
> I'm slightly annoyed by the extra amount of information that gets
> added to reorderbuffer.h for stuff that's only local to
> reorderbuffer.c, but that's not something new in this area, so..
We can actually remove the "lib/binaryheap.h" in this patch; I missed
that. There are no other uses of binaryheap in the file.
--
Heikki Linnakangas
Neon (https://neon.tech)
| From | Date | Subject | |
|---|---|---|---|
| Next Message | David Steele | 2024-04-10 23:36:11 | Re: post-freeze damage control |
| Previous Message | Heikki Linnakangas | 2024-04-10 22:46:04 | Re: [HACKERS] make async slave to wait for lsn to be replayed |