Hello Fabien,
I wrote:
> A guc parameter named 'checkpoint_partitions' is added.
> This can be set to 1, 2, 4, 8, or 16.
> Default is 16. (It is trivial at this present, I think.)
I've noticed that the behavior in 'checkpoint_partitions = 1' is not the
same as that of original 9.5alpha2.
Attached 'partitioned-checkpointing-v3.patch' fixed the bug, thus please use
it.
I'm sorry for that.
PS. The result graphs I sent was obtained using original 9.5alpha2, and thus
that results is unrelated to this bug.
As to 'partitioned checkpointing' case, the results shown in that graph
is probably worth than bug-fix version.
Best regards.
--
Takashi Horikawa
NEC Corporation
Knowledge Discovery Research Laboratories
> -----Original Message-----
> From: pgsql-hackers-owner@postgresql.org
> [mailto:pgsql-hackers-owner@postgresql.org] On Behalf Of Takashi Horikawa
> Sent: Saturday, September 12, 2015 11:50 AM
> To: Fabien COELHO
> Cc: pgsql-hackers@postgresql.org
> Subject: Re: [HACKERS] Partitioned checkpointing
>
> Hello Fabien,
>
> > I wanted to do some tests with this POC patch. For this purpose, it
would
> > be nice to have a guc which would allow to activate or not this feature.
> Thanks.
>
> > Could you provide a patch with such a guc? I would suggest to have the
> number
> > of partitions as a guc, so that choosing 1 would basically reflect the
> > current behavior.
> Sure.
> Please find the attached patch.
>
> A guc parameter named 'checkpoint_partitions' is added.
> This can be set to 1, 2, 4, 8, or 16.
> Default is 16. (It is trivial at this present, I think.)
>
> And some definitions are moved from bufmgr.h to xlog.h.
> It would be also trivial.
>
> Best regards.
> --
> Takashi Horikawa
> NEC Corporation
> Knowledge Discovery Research Laboratories