Re: Why Not MySQL? - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Why Not MySQL?
Date
Msg-id 7718.957367731@sss.pgh.pa.us
Whole thread Raw
In response to Re: Why Not MySQL?  ("Mitch Vincent" <mitch@huntsvilleal.com>)
List pgsql-hackers
"Mitch Vincent" <mitch@huntsvilleal.com> writes:
>> Could you be a little more specific on your performance issues ?

> Well, I'm just noticing that simple select queries are taking 3-5 seconds -
> on a table with 63 fields, 10000ish  rows of data. The ID fields are
> indexed, as well as several always-searched varchar() fields.

Hmm.  What does EXPLAIN show for the query plan?  You might also try
turning on execution stats (run psql with PGOPTIONS="-d2 -s" for
starters) to see what getrusage() can tell.  The results will be in the
postmaster log and might look like this:

StartTransactionCommand
query: SELECT usename, relname, relkind, relhasrules FROM pg_class, pg_user WHERE usesysid = relowner and ( relkind =
'r'OR relkind = 'i' OR relkind = 'S') and relname !~ '^pg_' and (relkind != 'i' OR relname !~ '^xinx') ORDER BY relname

ProcessQuery
! system usage stats:
!    0.083256 elapsed 0.040000 user 0.000000 system sec
!    [0.080000 user 0.020000 sys total]
!    12/1 [46/11] filesystem blocks in/out
!    0/0 [1/2] page faults/reclaims, 0 [0] swaps
!    0 [0] signals rcvd, 0/0 [2/2] messages rcvd/sent
!    8/5 [29/10] voluntary/involuntary context switches
! postgres usage stats:
!    Shared blocks:         18 read,          0 written, buffer hit rate = 94.29%
!    Local  blocks:          0 read,          0 written, buffer hit rate = 0.00%
!    Direct blocks:          0 read,          0 written
CommitTransactionCommand
        regards, tom lane


pgsql-hackers by date:

Previous
From: "Mitch Vincent"
Date:
Subject: Re: Why Not MySQL?
Next
From: The Hermit Hacker
Date:
Subject: Re: Why Not MySQL?