Nako le nako, ho hlaha mosebetsi oa ho batla lintlha tse amanang le sete sa linotlolo. ho fihlela re fumana palo e hlokahalang ea lirekoto.
Mohlala oa "bophelo ba sebele" ka ho fetisisa ke oa ho bonahatsa 20 mathata a khale ka ho fetisisa, ho thathamisitsoe lethathamong la basebetsi (mohlala, ka har'a karolo e le 'ngoe). Bakeng sa "li-dashboard" tse fapaneng tsa tsamaiso tse nang le likakaretso tse khutšoane tsa libaka tsa mosebetsi, sehlooho se ts'oanang se hlokahala hangata.
Sehloohong sena re tla sheba ts'ebetsong ho PostgreSQL ea tharollo ea "naive" bothateng bo joalo, "bohlale" le algorithm e rarahaneng haholo. "loop" ho SQL e nang le boemo ba ho tsoa ho data e fumanoeng, e ka ba molemo ka bobeli bakeng sa nts'etsopele e akaretsang le bakeng sa tšebeliso maemong a mang a tšoanang.
Ha re nke tlhahlobo ea data ho tloha . Ho thibela lirekoto tse bontšitsoeng hore li se ke tsa "tlola" nako le nako ha litekanyetso tse hlophisitsoeng li lumellana, atolosa index ea lihlooho ka ho eketsa senotlolo sa sehlooho. Ka nako e ts'oanang, sena se tla etsa hore e be se ikhethang 'me se re tiise hore tatellano ea ho hlopha ha e hlaka:
CREATE INDEX ON task(owner_id, task_date, id);
-- а старый - удалим
DROP INDEX task_owner_id_task_date_idx;Joalo ka ha ho utluoa, ho ngoliloe joalo
Taba ea pele, a re ke re rale mofuta o bonolo oa kopo, re fetise li-ID tsa libapali :
SELECT
*
FROM
task
WHERE
owner_id = ANY('{1,2,4,8,16,32,64,128,256,512}'::integer[])
ORDER BY
task_date, id
LIMIT 20;
Ke masoabi hanyane - re ile ra odara lirekoto tse 20 feela, empa Index Scan e ile ea re khutlisetsa eona 960 mela, eo ka nako eo hape e neng e lokela ho hlophisoa ... A re lekeng ho bala hanyenyane.
unnest + ARRAY
Taba ea pele e tla re thusa ke haeba re e hloka tse 20 feela tse hlophisitsoeng lirekoto, ebe u bala feela ha e fete 20 e hlophisitsoeng ka tatellano e tšoanang bakeng sa e 'ngoe le e 'ngoe senotlolo. Ho lokile, index e loketseng (owner_id, task_date, id) re na le.
Ha re sebeliseng mokhoa o ts'oanang oa ho hula le ho "hasalla likholomong" bohlokoa tafole rekoto, joalo ka . Re ka boela ra kenya kopo ea ho mena ka har'a sehlopha ho sebelisa ts'ebetso ARRAY():
WITH T AS (
SELECT
unnest(ARRAY(
SELECT
t
FROM
task t
WHERE
owner_id = unnest
ORDER BY
task_date, id
LIMIT 20 -- ограничиваем тут...
)) r
FROM
unnest('{1,2,4,8,16,32,64,128,256,512}'::integer[])
)
SELECT
(r).*
FROM
T
ORDER BY
(r).task_date, (r).id
LIMIT 20; -- ... и тут - тоже
Oh, ho betere haholo! 40% ka potlako le makhetlo a 4.5 ka tlase ho data Ke ile ka tlameha ho e bala.
Ho etsa lisebelisoa tsa lirekoto tsa tafole ka CTEEre ke le hlokomelise taba ea hore maemong a mang Boiteko ba ho sebetsa hang-hang le masimo a rekoto ka mor'a ho e batla ka subquery, ntle le "ho e phuthela" ka CTE, ho ka lebisa ho "ngatafatsa" InitPlan ho latela palo ea mafapha a tšoanang:
SELECT
((
SELECT
t
FROM
task t
WHERE
owner_id = 1
ORDER BY
task_date, id
LIMIT 1
).*);Result (cost=4.77..4.78 rows=1 width=16) (actual time=0.063..0.063 rows=1 loops=1)
Buffers: shared hit=16
InitPlan 1 (returns $0)
-> Limit (cost=0.42..1.19 rows=1 width=48) (actual time=0.031..0.032 rows=1 loops=1)
Buffers: shared hit=4
-> Index Scan using task_owner_id_task_date_id_idx on task t (cost=0.42..387.57 rows=500 width=48) (actual time=0.030..0.030 rows=1 loops=1)
Index Cond: (owner_id = 1)
Buffers: shared hit=4
InitPlan 2 (returns $1)
-> Limit (cost=0.42..1.19 rows=1 width=48) (actual time=0.008..0.009 rows=1 loops=1)
Buffers: shared hit=4
-> Index Scan using task_owner_id_task_date_id_idx on task t_1 (cost=0.42..387.57 rows=500 width=48) (actual time=0.008..0.008 rows=1 loops=1)
Index Cond: (owner_id = 1)
Buffers: shared hit=4
InitPlan 3 (returns $2)
-> Limit (cost=0.42..1.19 rows=1 width=48) (actual time=0.008..0.008 rows=1 loops=1)
Buffers: shared hit=4
-> Index Scan using task_owner_id_task_date_id_idx on task t_2 (cost=0.42..387.57 rows=500 width=48) (actual time=0.008..0.008 rows=1 loops=1)
Index Cond: (owner_id = 1)
Buffers: shared hit=4"
InitPlan 4 (returns $3)
-> Limit (cost=0.42..1.19 rows=1 width=48) (actual time=0.009..0.009 rows=1 loops=1)
Buffers: shared hit=4
-> Index Scan using task_owner_id_task_date_id_idx on task t_3 (cost=0.42..387.57 rows=500 width=48) (actual time=0.009..0.009 rows=1 loops=1)
Index Cond: (owner_id = 1)
Buffers: shared hit=4
Tlaleho e tšoanang e ne e "sheba holimo" ka makhetlo a 4 ... Ho fihlela PostgreSQL 11, boitšoaro bona bo etsahala kamehla, 'me tharollo ke ho e "phuthela" ka CTE, e leng moeli o feletseng oa optimizer liphetolelong tsena.
Recursive accumulator
Phetolelong e fetileng, ka kakaretso rea bala 200 mela molemong oa tse hlokahalang 20. Eseng 960, empa le ho feta - na hoa khoneha?
A re lekeng ho sebelisa tsebo eo re e hlokang palo yohle 20 litlaleho. Ke hore, re tla pheta ho bala ha data feela ho fihlela re fihla chelete eo re e hlokang.
Mohato oa 1: Lethathamo la ho Qala
Ho hlakile hore lethathamo la rona la "sepheo" sa lirekoto tse 20 le lokela ho qala ka lirekoto tsa "pele" bakeng sa e 'ngoe ea linotlolo tsa owner_id. Ka hona, pele re tla fumana tse joalo “pele haholo” bakeng sa senotlolo ka seng 'me u e kenye lethathamong, u e hlophise ka tatellano eo re e batlang - (task_date, id).
Mohato oa 2: Fumana "tse latelang".
Joale haeba re nka karolo ea pele lethathamong la rona ebe re qala “step” ho feta ho latela index ho boloka senotlolo sa owner_id, joale lirekoto tsohle tse fumanoeng ke tsona tse latelang khethong e hlahisoang. Ehlile, feela ho fihlela re tšela senotlolo sa butt keno ea bobeli lethathamong.
Haeba ho bonahala eka re "tšetse" tlaleho ea bobeli, joale keno ea ho qetela e baliloeng e lokela ho eketsoa lethathamong ho fapana le ea pele (ka mong_id ea tšoanang), ka mor'a moo re hlophisa lethathamo hape.
Ke hore, kamehla re fumana hore lenane ha le na ho kena ho feta e le 'ngoe bakeng sa linotlolo ka' ngoe (haeba lipehelo li fela 'me re sa "tšela", joale lenane la pele le hlahang lethathamong le tla nyamela feela 'me ha ho letho le tla eketsoa. ), le bona ka mehla e hlophiloe ka tatellano e nyolohang ea senotlolo sa kopo (task_date, id).
Mohato oa 3: filthara 'me u "atolose" lirekoto
Mela e meng ea khetho ea rona e pheta-phetoang, tse ling li rekotoa rv li kopitsoe - pele re fumana joalo ka "ho tšela moeli oa keno ea 2 ea lenane", ebe re e nkela sebaka sa 1 lethathamong. Kahoo ketsahalo ea pele e hloka ho hlophisoa.
Potso ea ho qetela e tšabehang
WITH RECURSIVE T AS (
-- #1 : заносим в список "первые" записи по каждому из ключей набора
WITH wrap AS ( -- "материализуем" record'ы, чтобы обращение к полям не вызывало умножения InitPlan/SubPlan
WITH T AS (
SELECT
(
SELECT
r
FROM
task r
WHERE
owner_id = unnest
ORDER BY
task_date, id
LIMIT 1
) r
FROM
unnest('{1,2,4,8,16,32,64,128,256,512}'::integer[])
)
SELECT
array_agg(r ORDER BY (r).task_date, (r).id) list -- сортируем список в нужном порядке
FROM
T
)
SELECT
list
, list[1] rv
, FALSE not_cross
, 0 size
FROM
wrap
UNION ALL
-- #2 : вычитываем записи 1-го по порядку ключа, пока не перешагнем через запись 2-го
SELECT
CASE
-- если ничего не найдено для ключа 1-й записи
WHEN X._r IS NOT DISTINCT FROM NULL THEN
T.list[2:] -- убираем ее из списка
-- если мы НЕ пересекли прикладной ключ 2-й записи
WHEN X.not_cross THEN
T.list -- просто протягиваем тот же список без модификаций
-- если в списке уже нет 2-й записи
WHEN T.list[2] IS NULL THEN
-- просто возвращаем пустой список
'{}'
-- пересортировываем словарь, убирая 1-ю запись и добавляя последнюю из найденных
ELSE (
SELECT
coalesce(T.list[2] || array_agg(r ORDER BY (r).task_date, (r).id), '{}')
FROM
unnest(T.list[3:] || X._r) r
)
END
, X._r
, X.not_cross
, T.size + X.not_cross::integer
FROM
T
, LATERAL(
WITH wrap AS ( -- "материализуем" record
SELECT
CASE
-- если все-таки "перешагнули" через 2-ю запись
WHEN NOT T.not_cross
-- то нужная запись - первая из спписка
THEN T.list[1]
ELSE ( -- если не пересекли, то ключ остался как в предыдущей записи - отталкиваемся от нее
SELECT
_r
FROM
task _r
WHERE
owner_id = (rv).owner_id AND
(task_date, id) > ((rv).task_date, (rv).id)
ORDER BY
task_date, id
LIMIT 1
)
END _r
)
SELECT
_r
, CASE
-- если 2-й записи уже нет в списке, но мы хоть что-то нашли
WHEN list[2] IS NULL AND _r IS DISTINCT FROM NULL THEN
TRUE
ELSE -- ничего не нашли или "перешагнули"
coalesce(((_r).task_date, (_r).id) < ((list[2]).task_date, (list[2]).id), FALSE)
END not_cross
FROM
wrap
) X
WHERE
T.size < 20 AND -- ограничиваем тут количество
T.list IS DISTINCT FROM '{}' -- или пока список не кончился
)
-- #3 : "разворачиваем" записи - порядок гарантирован по построению
SELECT
(rv).*
FROM
T
WHERE
not_cross; -- берем только "непересекающие" записи
Kahoo, re e rekisitsoe 50% ea data e baloang bakeng sa 20% ea nako ea ts'ebetso. Ke hore, haeba u na le mabaka a ho lumela hore ho bala ho ka nka nako e telele (mohlala, data hangata ha e eo ka har'a cache, 'me u tlameha ho ea ho disk bakeng sa eona), joale ka tsela ena u ka itšetleha ka ho fokolang ho bala. .
Leha ho le joalo, nako ea ho bolaoa e bile betere ho feta khetho ea pele ea "naive". Empa ke efe ea likhetho tse 3 tseo u ka li sebelisang ho uena.
Source: www.habr.com