On 06.03.2025 16:30, Matthias van de Meent wrote:
On Thu, 6 Mar 2025 at 14:18, Alena Rybakina <a.rybak...@postgrespro.ru> wrote:
Hi! I got a query plan with a strange number of rows. Could you please
help me understand it?
To be honest I can't understand why 0.50 number of rows here?
Because the scan matched only ~(500 rows over 999 iterations = 500/999
~=) 0.50 rows for every loop, on average, for these plan nodes:
-> Nested Loop (actual rows=0.50 loops=999)
-> Seq Scan on tb (actual rows=0.50 loops=999)
And for this, it was 500 rows total in 1000 iterations, which also
rounds to 0.50:
SubPlan 2
-> Result (actual rows=0.50 loops=1000)
One-Time Filter: ((ta1.id < 1000) AND (InitPlan 1).col1)
As of ddb17e38 (and its follow-up 95dbd827), we display fractional
rows-per-loop, with 2 digits of precision, rather than a rounded
integer. This allows a user to distinguish plan nodes with 0.49
rows/loop and 0.01 rows/loop, and that can help inform the user about
how to further optimize their usage of indexes and other optimization
paths.
Thanks for the explanation. Now I understand.
--
Regards,
Alena Rybakina
Postgres Professional
