Horton

Re: Default setting for enable_hashagg_disk

Alvaro Herrera <alvherre@2ndquadrant.com>

View thread View on postgr.es
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Peter Geoghegan <pg@bowt.ie>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Stephen Frost <sfrost@snowman.net>, Jeff Davis <pgsql@j-davis.com>, Andres Freund <andres@anarazel.de>, Bruce Momjian <bruce@momjian.us>, Robert Haas <robertmhaas@gmail.com>, David Rowley <dgrowleyml@gmail.com>, Justin Pryzby <pryzby@telsasoft.com>, Melanie Plageman <melanieplageman@gmail.com>, Tomas Vondra <tomas.vondra@2ndquadrant.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2020-07-10T21:10:26Z
Lists: pgsql-hackers, pgsql-docs 
On 2020-Jul-10, Peter Geoghegan wrote:

> * The maximum allowable value is 100.0, to protect users from
> accidentally setting hash_mem_multiplier to a value intended to work
> like a work_mem-style KB value (you can't provide an absolute value
> like that directly). This maximum is absurdly high.
> 
> I think that it's possible that a small number of users will find it
> useful to set the value of hash_mem_multiplier as high as 5.0. That is
> a very aggressive value, but one that could still make sense with
> certain workloads.

I'm not sure about this bit; sounds a bit like what has been qualified
as "nannyism" elsewhere.  Suppose I want to give a hash table 2GB of
memory for whatever reason.  If my work_mem is default (4MB) then I
cannot possibly achieve that without altering both settings.

So I propose that maybe we do want a maximum value, but if so it should
be higher than what you propose.  I think 10000 is acceptable in that it
doesn't get in the way.

Another point is that if you specify a unit for the multiplier (which is
what users are likely to do for larger values), it'll fail anyway, so
I'm not sure this is such terrible a problem.

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

Commits

  1. Add hash_mem_multiplier GUC.

  2. HashAgg: use better cardinality estimate for recursive spilling.

  3. Remove hashagg_avoid_disk_plan GUC.

  4. Doc fixup for hashagg_avoid_disk_plan GUC.

  5. Rework HashAgg GUCs.

  6. Disk-based Hash Aggregation.

  7. Implement partition-wise grouping/aggregation.

  8. Defer creation of partially-grouped relation until it's needed.