Re: Patch set under development to add usage reporting.

[Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>]

Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> On lör, 2009-10-31 at 13:37 -0400, John Murtari wrote:
>> ...  We also had
>> developed the "SHOW STATISTICS" syntax for DB version that don't
>> support INFORMATION SCHEMA.

> I don't really get the point of the SHOW STATISTICS command.  There is
> already a command whose purpose is to retrieve data in tabular form,
> namely SELECT.

I presume we need not worry about that, since the SQL committee are
certainly not going to standardize something that's only there for
DBs that don't support information_schema.

Actually, the idea of standardizing anything at all in this area seems
pretty bogus.  The events that are interesting to measure are below the
semantic level of the standard --- for instance, how are you going to
count index searches, when the standard doesn't even recognize the
existence of indexes?  Let alone things like buffer cache hits, which is
a concept that might not exist at all in some implementations.  I think
we can safely assume that the proposed standardization effort will go
nowhere, and just look at whether this is interesting for Postgres.

			regards, tom lane