You are lying to us about how those queries were posed to Postgres
(and no I don't feel a need to explain how I know).
Sorry. The "lying" was not intended as explained in my reply to Heikku.
Thanks for the tips anyways.
---(end of broadcast)---
T
There's something odd about that plan. It's doing both a seq scan and a
bitmap scan on "items", but I can't see stats table being mentioned
anywhere.
Huh? Aaah, sorry. I made a major search/replace-refactoring (that
obviously went wrong) on all open files in the editor before posting to
this
cluster <[EMAIL PROTECTED]> writes:
> Queries and output from EXPLAIN ANALYZE can be seen here with some
> syntax highlighting:
> http://rafb.net/p/BJIW4p69.html
You are lying to us about how those queries were posed to Postgres
(and no I don't feel a need to explain how I know). In future p
cluster wrote:
SELECT keyId, sortNum, count(1)
FROM stats s, items i
WHERE s.keyId = i.keyId AND i.sortNum > 123
GROUP BY i.keyId, i.sortNum
ORDER BY i.sortNum
LIMIT 50
Limit (cost=3281.72..3281.84 rows=50 width=16) (actual
time=435.838..436.043 rows=50 loops=1)
InitPlan
-
I have two small queries which are both very fast to evaluate
separately. The first query, "Query 1", calculates some statistics and
the the second query, "Query 2", finds a subset of relevant keys.
When combined into a single query which calculates statistics from only
the subset of relevant ke
