On Thu, Jan 31, 2013 at 2:40 PM, Alvaro Herrera
<alvherre@2ndquadrant.com> wrote:
> Alvaro Herrera escribió:
>
>> Okay, here's a patch along these lines. I haven't considered Jim's
>> suggestion downthread about discounting dead tuples from relpages; maybe
>> we can do that by subtracting the pages attributed to dead ones,
>> estimating via tuple density (reltuples/relpages).
>
> Patch attached.
This strikes me as too clever by half. You've introduced the concept
of a "Browne strength" (apparently named for Christopher Browne) and
yet you haven't even bothered to add a comment explaining the meaning
of the term, let along justifying the choice of that formula rather
than any other. I don't want to dog this proposal to death, because
surely we can do better than the status quo here, but adopting the
first formula someone proposed without any analysis of whether it does
the right thing cannot possibly be the correct decision process.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company