MIME-Version: 1.0
References: <20230828115252.c1b018605b9a0756a30c3382@sraoss.co.jp>
 <20230828160530.adde1e20f257d7d345989163@sraoss.co.jp> <CACJufxEoCCJE1vntJp1SWjen8vBUa3vZLgL=swPwar4zim976g@mail.gmail.com>
 <20230902.204634.955758704959569058.t-ishii@sranhm.sra.co.jp>
 <CACJufxFjankFQDNppOfqCTpY=zW4Q0+2WCmKjT95kggiT978Lw@mail.gmail.com>
 <CAHut+PsDpBTxZ7bLhko7_E-C7khMhoNJcriNQ_p_gWjADn01vg@mail.gmail.com>
 <20240123162327.c2803162619dd7634cca0b6c@sraoss.co.jp> <20240304115846.2275fb44fd904e8789d43590@sraoss.co.jp>
 <20240329234700.73ff2e28c9248d29f8fa6a66@sraoss.co.jp> <20240331225931.712683cecb26862b73b2b822@sraoss.co.jp>
 <20240702170311.1ddb417759a48ff12c555b92@sranhm.sraoss.co.jp.sranhm>
 <20240711132357.fe3f78c184cfa99159208178@sranhm.sraoss.co.jp>
 <CALdSSPhj1H1NS7QiYkSQNCksPCwjtLcyt3==evgkBX1SrKyVdQ@mail.gmail.com>
 <CALdSSPip9ruUoQMmsD_hQ0xY72qB=_jB-ayHeUWUH-dd0MB60A@mail.gmail.com>
 <20240730142420.34a9ad7c249aecde88cd45fb@sraoss.co.jp> <CALdSSPiBBBCHxVtg+X6OdBkJPGYOvLf1hST4MgBgRKZh0Xddyw@mail.gmail.com>
In-Reply-To: <CALdSSPiBBBCHxVtg+X6OdBkJPGYOvLf1hST4MgBgRKZh0Xddyw@mail.gmail.com>
From: Kirill Reshke <reshkekirill@gmail.com>
Date: Tue, 20 Aug 2024 12:06:24 +0500
Message-ID: <CALdSSPg1mPiZRXjFEsz3h_5Jue0rq0w9BJzBrcVtwz5J29d_3w@mail.gmail.com>
Subject: Re: Incremental View Maintenance, take 2
To: Yugo NAGATA <nagata@sraoss.co.jp>
Cc: Peter Smith <smithpb2250@gmail.com>, jian he <jian.universality@gmail.com>, 
	Tatsuo Ishii <ishii@sraoss.co.jp>, pgsql-hackers@postgresql.org
Content-Type: text/plain; charset="UTF-8"
Archived-At: <https://www.postgresql.org/message-id/CALdSSPg1mPiZRXjFEsz3h_5Jue0rq0w9BJzBrcVtwz5J29d_3w%40mail.gmail.com>
Precedence: bulk

On Tue, 20 Aug 2024 at 02:14, Kirill Reshke <reshkekirill@gmail.com> wrote:
>
>
> == Major suggestions.
>
> 1) At first glance, working with this IVM/IMMV infrastructure feels
> really unintuitive about what servers actually do for query execution.
> I do think It will be much better for user experience to add more
> EXPLAIN about IVM work done inside IVM triggers. This way it is much
> clearer which part is working slow, so which index should be created,
> etc.
>
> 2) The kernel code for IVM lacks possibility to be extended for
> further IVM optimizations. The one example is foreign key optimization
> described here[1]. I'm not saying we should implement this within this
> patchset, but we surely should pave the way for this. I don't have any
> good suggestions for how to do this though.
>
> 3) I don't really think SQL design is good. CREATE [INCREMENTAL] M.V.
> is too ad-hoc. I would prefer CREATE M.V. with (maintain_incr=true).
> (reloption name is just an example).
> This way we can change regular M.V. to IVM and vice versa via ALTER
> M.V. SET *reloptions* - a type of syntax that is already present in
> PostgreSQL core.
>

One little follow-up here. Why do we do prepstate visibility the way
it is done? Can we instead export the snapshot in BEFORE trigger, save
it somewhere and use it after?

-- 
Best regards,
Kirill Reshke