Re: progress report for ANALYZE - Mailing list pgsql-hackers

From Alvaro Herrera
Subject Re: progress report for ANALYZE
Date
Msg-id 20191127030014.GD4161@alvherre.pgsql
Whole thread Raw
In response to Re: progress report for ANALYZE  (Tatsuro Yamada <tatsuro.yamada.tf@nttcom.co.jp>)
List pgsql-hackers
On 2019-Nov-27, Amit Langote wrote:

> On Tue, Nov 26, 2019 at 9:22 PM Alvaro Herrera <alvherre@2ndquadrant.com> wrote:
> >
> > On 2019-Nov-26, Tatsuro Yamada wrote:
> >
> > > > I wonder whether we need the total number of ext stats on
> > > > pg_stat_progress_analyze or not. As you might know, there is the same
> > > > counter on pg_stat_progress_vacuum and pg_stat_progress_cluster.
> > > > For example, index_vacuum_count and index_rebuild_count.
> > >
> > > Would it be better to add the total number column to these views? :)
> >
> > Yeah, I think it would be good to add that.
> 
> Hmm, does it take that long to calculate ext stats on one column?  The
> reason it's worthwhile to have index_vacuum_count,
> index_rebuild_count, etc. is because it can take very long for one
> index to get vacuumed/rebuilt.

Yes, it's noticeable.  It's not as long as building an index, of course,
but it's a long enough fraction of the total analyze time that it should
be reported.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services



pgsql-hackers by date:

Previous
From: Amit Kapila
Date:
Subject: Re: [HACKERS] Block level parallel vacuum
Next
From: Kyotaro Horiguchi
Date:
Subject: Re: ERROR: attribute number 6 exceeds number of columns 5