Re: Patch: add timing of buffer I/O requests

From: Peter Geoghegan <peter(at)2ndquadrant(dot)com>
To: Robert Haas <robertmhaas(at)gmail(dot)com>
Cc: Ants Aasma <ants(at)cybertec(dot)at>, Greg Smith <greg(at)2ndquadrant(dot)com>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: Patch: add timing of buffer I/O requests
Date: 2012-04-10 14:11:47
Message-ID: CAEYLb_V1d+oDCiefqFQBztaOgWtKdBnZkS9dqvwMYCMpso+gvg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 10 April 2012 14:33, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> So, should we make the new columns exposed by pg_stat_statements use
> milliseconds, so that the block read/write timings are everywhere in
> milliseconds, or should we keep them as a float8, so that all the
> times exposed by pg_stat_statements use float8?

I believe that we should keep them as float8, on the basis that a user
is more likely to generalise from total_time's format (when writing a
script to query the view of whatever) than from that of
pg_stat_database.

A part of me would like to change the view definitions so that all the
columns are strongly typed (i.e. all these values would be intervals).
I realise that that isn't practical though.

--
Peter Geoghegan       http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training and Services

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Andrew Dunstan 2012-04-10 14:15:37 Re: [JDBC] Regarding GSoc Application
Previous Message Tom Lane 2012-04-10 14:06:29 Re: Patch: add timing of buffer I/O requests