query uses index but takes too much time?

hello to all,

i would like your advice on the following matter. i have a table with 150
million rows. there are some indexes on this table but the one that is
really important is one that has 3 columns (a,b,c). one application
constantly makes queries and the query planner uses this index to narrow
down the final set of results. so usually from 150 millions, when the 3
conditions have been applied, the remaining rows to be checked are about
20-300. So these queries are very fast, and take from 10-100 ms usually.
There is a special case where these 3 conditions narrow down the final set
to 15.000 rows so the server must check all these rows. The result is that
the query takes around 1 minute to complete. Is that a normal time for the
execution of the query?

i know that most of you will send me the link with the guide to reporting
slow queries but that's not the point at the moment. i am not looking for a
specific answer why this is happening.
i just want to know if that seems strange to more people than just me and if
i should look into that.

but if for the above you need to have a clearer picture of the server then:
-red hat 5.6
-32 cores,
-96GB ram
-fiber storage (4GBps)
-postgresql 9.0.5
-shared_buffers : 25 GB
-not i/o bound (too many disks, different partitions for backup, archives,
xlogs, indexes)
-not cpu bound (the cpu util was about 5% when i performed the tests)
-the query planner values on postgresql.conf are the default
-i also performed the tests on the hot-standby with the same results
-the query plan is the correct one, indicating that it should use the
correct index
-i forced index_scan to off and then it used bitmap heap scan with similar
results.
-i forced bitmap heap scan to off and then it did a seq scan

any ideas? thx in advance for your insight

--
View this message in context: http://postgresql.1045698.n5.nabble.com/query-uses-index-but-takes-too-much-time-tp5020742p5020742.html
Sent from the PostgreSQL - performance mailing list archive at Nabble.com.