Re: Re: patch review : Add ability to constrain backend temporary file space
Mark Kirkwood <mark.kirkwood@catalyst.net.nz>
From: Mark Kirkwood <mark.kirkwood@catalyst.net.nz>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Robert Haas <robertmhaas@gmail.com>, Cédric Villemain <cedric.villemain.debian@gmail.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2011-06-02T22:25:57Z
Lists: pgsql-hackers
On 03/06/11 02:36, Tom Lane wrote: > Robert Haas<robertmhaas@gmail.com> writes: >> So I'm not sure work_disk is a great name. > I agree. Maybe something along the lines of "temp_file_limit"? > > Also, once you free yourself from the analogy to work_mem, you could > adopt some more natural unit than KB. I'd think MB would be a practical > unit size, and would avoid (at least for the near term) the need to make > the parameter a float. > I agree, and I like your name suggestion (and also agree with Robert that 'work_mem' was not such a great name). I'd be quite happy to use MB (checks guc.h) but looks like we don't have a GUC_UNIT_MB, not sure if adding it would be an issue (suggests on a suitable mask if people think it is a great idea). Cheers Mark