> Can you send `EXPLAIN (analyze, buffers)` for your query instead?
> It'll show exactly what's going on.
GroupAggregate (cost=303425.31..339014.43 rows=136882 width=48) (actual
time=4708.181..6688.699 rows=287268 loops=1)
Buffers: shared read=23899, temp read=30974 written=30974
-> Sort (cost=303425.31..306847.34 rows=1368812 width=48) (actual
time=4708.170..5319.429 rows=1368744 loops=1)
Sort Key: co_id, client_id, doc_no,
Sort Method: external merge Disk: 80304kB
Buffers: shared read=23899, temp read=30974 written=30974
-> Seq Scan on ddetail (cost=0.00..37587.12 rows=1368812 width=48)
(actual time=0.122..492.964 rows=1368744 loops=1)
Buffers: shared read=23899
Total runtime: 6708.244 ms
My initial attempt was this (this is what I actually need):
SELECT co_id , client_id , doc_no , line_id , batch_no , sum( amount )
FROM ddetail
GROUP BY co_id , client_id , doc_no , line_id , batch_no ;
but I removed column batch_no from the query because I thought this was the
cause of the problem ( batch_no is not part of my PK ).
Thanks
--
View this message in context:
http://postgresql.1045698.n5.nabble.com/Performance-issue-index-not-used-on-GROUP-BY-tp5816702p5816706.html
Sent from the PostgreSQL - performance mailing list archive at Nabble.com.