PostgresOpen 2019 Explain Plans And You

[@kunalmandalia1165]

Excellent talk. The speaker is keen for the audience to learn as opposed to appearing clever and that shines through.

[@kundan621]

Thanks for sharing. Excellent presentation and great content.

[@india_is_greatest]

by far the best talk I listened so far about Postgres

[@deryckhenson4353]

Very good talk. Wish resolution was higher couldn't read much on the screen :\

[@YogeshJadhav96]

Great Talk.

[@SARANSASIDHARAN1]

Excellent talk!

[@user-fr2fd4vj4p]

Thanks. It was very helpful

[@mourinhoalex1992]

What is the execution order of this plan: Sort (cost=87.71..88.75 rows=418 width=40) Sort Key: language.name DESC -> Hash Join (cost=1.14..69.51 rows=418 width=40) Hash Cond: (film.language_id = language.language_id) -> Seq Scan on film (cost=0.00..66.50 rows=418 width=21) Filter: (rating = ANY ('{R,PG-13}'::mpaa_rating[])) -> Hash (cost=1.06..1.06 rows=6 width=25) -> Seq Scan on language (cost=0.00..1.06 rows=6 width=25) As I understand from the talk, it should be: 1. The nodes that have the startup cost equals to 0 will be the nodes that gets run first, so these 2 nodes runs in parallel: Seq Scan on film, Seq Scan on language 2. Then this node gets run: Hash (cost=1.06..1.06 rows=6 width=25) 3. Then this node: Hash Join 4. Finally: Sort Do I understand it correctly ? If not, then what is the correct execution order of the nodes in this plan ?