зеркало из
https://github.com/iharh/notes.git
synced 2025-10-31 05:36:08 +02:00
28 строки
1.7 KiB
Plaintext
28 строки
1.7 KiB
Plaintext
2024
|
|
Digitalize - PG Optimization Basics 1:54:00 of 2:07:51
|
|
https://www.youtube.com/watch?v=gA3A_epB3So
|
|
! 13:00 - from index pages go into buffer-cache in RAM
|
|
! 37:00 - select ... from employee e join contacts ec using(employee_id) -- on e.employee_id = ec.employee_id
|
|
! 39:00 - data access methods: seq-scan, idx-scan(with going into table), index-only-scan(without going into table), bitmap-heap-scan
|
|
! 45:00 - Seq Scan on employee (cost=0.00..173528.84 rows ...)
|
|
! 0.00 - cost of first raw obtaining, 173528.84 - cost of all rows obtaining
|
|
! 47:00 select relpages, current_setting('seq_page_cost'), relpages*current_setting('seq_page_cost')::int as total
|
|
! from pg_class where relname='employee';
|
|
! relpages: 73 530
|
|
! current_setting: 1
|
|
! total: 73 530
|
|
! 53:00 selectivity (0.0 .. 1.0) - percentage/proportion of chosen rows
|
|
! idx is effectively used only on low-selectivity, otherwise - seq scan
|
|
! 1:08:00 - vacuum analyze - for updating statistics
|
|
! 1:17:00 - create extension pg_stat_statements; alter system set shared_preload_libraries = 'pg_stat_statements';
|
|
! 1:18:00 - other perf-related settings !!!
|
|
! recommendation - https://www.youtube.com/playlist?list=PLaFqU3KCWw6JW80WBHPOe-SMJD2NOjmge
|
|
JPoint - Zhilin - PG Can Make a Pain 0:00 of 45:49
|
|
https://www.youtube.com/watch?v=pTWfjyeQ5Ks
|
|
https://jokerconf.com/talks/0bd645ad3a0f48c292547db572a1c7eb
|
|
WorkSolutions - Kostenko - Postgres World Optimization Misteries ru 0:00 of 32:50
|
|
https://www.youtube.com/watch?v=MeTDmLyYD9k
|
|
2023
|
|
Tenzor - Backend School of p9
|
|
https://www.youtube.com/playlist?list=PLt0vzWoDuwcTdFnp-QWtx2yEvKMDlPw7l
|