Marmar, hawsha raadinta xogta la xidhiidha iyadoo la adeegsanayo furayaal ayaa soo baxda. ilaa aan ka helno tirada guud ee diiwaanada loo baahan yahay.
Tusaalaha "nolosha dhabta ah" ee ugu badan waa in la muujiyo 20-ka dhibaato ee ugu da'da weyn, ku taxan liiska shaqaalaha (tusaale ahaan, hal qayb gudahood). Maaraynta kala duwan ee "dashboards" oo leh koobab kooban oo meelaha shaqada ah, mawduuc la mid ah ayaa loo baahan yahay marar badan.
Maqaalkan waxaan ku eegi doonaa hirgelinta PostgreSQL ee xalinta "naive" ee dhibaatadan oo kale, "ka caqli badan" iyo algorithm aad u adag. "loop" gudaha SQL oo leh xaalad ka bixitaan xogta la helay, kaas oo waxtar u yeelan kara horumarka guud iyo isticmaalka xaaladaha kale ee la midka ah labadaba.
Aynu ka soo qaadanno xogta tijaabada ah
CREATE INDEX ON task(owner_id, task_date, id);
-- Π° ΡΡΠ°ΡΡΠΉ - ΡΠ΄Π°Π»ΠΈΠΌ
DROP INDEX task_owner_id_task_date_idx;
Sida loo maqlo ayaa loo qoray
Marka hore, aan sawirno nooca ugu fudud ee codsiga, anagoo gudbinayna aqoonsiga hawlwadeenada
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;
In yar oo murugo leh - waxaanu dalbanay kaliya 20 diiwaan, laakiin Index Scan ayaa nagu soo celiyay 960 sadar, kaas oo sidoo kale lahaa in la kala saaro... Aan isku dayno inaan wax yar akhrino.
unnest + ARRAY
Tixgelinta ugu horreysa ee ina caawin doonta waa haddii aan u baahanahay kaliya 20 la soocay diiwaanada, ka dibna kaliya akhri wax aan ka badnayn 20 ayaa loo kala soocay si isku mid ah mid kasta furaha. Wacan index ku habboon (owner_id, task_date, id) waan haynaa.
Aynu isticmaalno hab isku mid ah soo saarista iyo "ku faafinta tiirarka" diiwaanka miiska aasaasiga ah, sida ku jirta ARRAY()
:
WITH T AS (
SELECT
unnest(ARRAY(
SELECT
t
FROM
task t
WHERE
owner_id = unnest
ORDER BY
task_date, id
LIMIT 20 -- ΠΎΠ³ΡΠ°Π½ΠΈΡΠΈΠ²Π°Π΅ΠΌ ΡΡΡ...
)) 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; -- ... ΠΈ ΡΡΡ - ΡΠΎΠΆΠ΅
Oh, mar hore aad uga wanaagsan! 40% dhakhso badan iyo 4.5 jeer ka yar xogta Waxay ahayd inaan akhriyo.
Qalabaynta diiwaannada miiska iyada oo loo marayo CTEAan ku soo jiido dareenkaaga xaqiiqda xaaladaha qaarkood Isku dayga in isla markiiba lagala shaqeeyo goobaha diiwaanka ka dib markaad ka raadiso subquery, iyada oo aan "ku duubin" CTE, waxay horseedi kartaa "ku dhufo" InitPlan oo u dhiganta tirada goobahan:
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
Isla diiwaanka ayaa "kor loo eegay" 4 jeer... Ilaa PostgreSQL 11, habdhaqankani si joogto ah ayuu u dhacaa, xalkuna waa in lagu "duubo" CTE, taas oo xad buuxda u ah hagaajinta noocyadan.
Accumulator soo noqnoqda
In version hore, guud ahaan waanu akhrinay 200 sadar danta loo baahan yahay 20. Ma aha 960, laakiin xataa ka yar - suurto gal ah?
Aan isku dayno inaan ka faa'ideyno aqoonta aan u baahanahay wadarta 20 diiwaanada. Taasi waa, waxaanu ku celcelin doonaa akhrinta xogta kaliya ilaa aan ka gaarno qadarka aan u baahanahay.
Talaabada 1: Liiska Bilaabida
Sida iska cad, liiskayaga "bartilmaameedka" ee 20ka diiwaan waa inay ku bilowdaan diiwaanada "koowaad" ee mid ka mid ah furayaashayada_id. Sidaa darteed, marka hore waxaan heli doonaa sida "aad u horraysa" mid kasta oo ka mid ah furayaasha oo ku dar liiska, una kala sooc habka aan rabno - (hawl_date, id).
Talaabada 2: Soo hel gelinta "xiga".
Hadda haddii aan ka soo qaadanno gelitaanka ugu horreeya liiskayaga oo aan bilowno "Tallaabo" oo ka sii socota tusmada ilaalinta furaha owner_id, ka dib dhammaan diiwaanada la helay waa kuwa ku xiga xulashada natiijada. Dabcan, kaliya ilaa aan ka gudubno furaha dabada gelitaanka labaad ee liiska.
Haddii ay soo baxdo inaan "ka gudubnay" rikoodhka labaad, markaa akhrinta u dambaysa waa in lagu daraa liiska halkii lagu dari lahaa kii hore (oo leh isla owner_id), ka dib waxaan dib u kala soocnay liiska mar kale.
Taasi waa, waxaan had iyo jeer helnaa in liisku uusan lahayn wax ka badan hal gelitaan mid kasta oo ka mid ah furayaasha (haddii gelintadu dhammaato oo aynaan "gudbin", markaa gelitaanka ugu horreeya ee liiska ayaa si fudud u baaba'aya waxna laguma dari doono. ), iyo iyaga had iyo jeer kala soocida sida u koraya ee furaha codsiga (task_date, id).
Talaabada 3: filter iyo "ballaari" diiwaanada
Qaar ka mid ah safafka xulashadayada soo noqnoqda, diiwaanada qaar rv
waa la labanlaabay - marka hore waxaan helnaa sida "ka gudubka xadka gelitaanka 2aad ee liiska", ka dibna u beddelo sida 1aad ee liiska. Markaa dhacdada ugu horreysa waxay u baahan tahay in la sifeeyo.
Weydiinta kama dambaysta ah ee cabsida leh
WITH RECURSIVE T AS (
-- #1 : Π·Π°Π½ΠΎΡΠΈΠΌ Π² ΡΠΏΠΈΡΠΎΠΊ "ΠΏΠ΅ΡΠ²ΡΠ΅" Π·Π°ΠΏΠΈΡΠΈ ΠΏΠΎ ΠΊΠ°ΠΆΠ΄ΠΎΠΌΡ ΠΈΠ· ΠΊΠ»ΡΡΠ΅ΠΉ Π½Π°Π±ΠΎΡΠ°
WITH wrap AS ( -- "ΠΌΠ°ΡΠ΅ΡΠΈΠ°Π»ΠΈΠ·ΡΠ΅ΠΌ" record'Ρ, ΡΡΠΎΠ±Ρ ΠΎΠ±ΡΠ°ΡΠ΅Π½ΠΈΠ΅ ΠΊ ΠΏΠΎΠ»ΡΠΌ Π½Π΅ Π²ΡΠ·ΡΠ²Π°Π»ΠΎ ΡΠΌΠ½ΠΎΠΆΠ΅Π½ΠΈΡ 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 -- ΡΠΎΡΡΠΈΡΡΠ΅ΠΌ ΡΠΏΠΈΡΠΎΠΊ Π² Π½ΡΠΆΠ½ΠΎΠΌ ΠΏΠΎΡΡΠ΄ΠΊΠ΅
FROM
T
)
SELECT
list
, list[1] rv
, FALSE not_cross
, 0 size
FROM
wrap
UNION ALL
-- #2 : Π²ΡΡΠΈΡΡΠ²Π°Π΅ΠΌ Π·Π°ΠΏΠΈΡΠΈ 1-Π³ΠΎ ΠΏΠΎ ΠΏΠΎΡΡΠ΄ΠΊΡ ΠΊΠ»ΡΡΠ°, ΠΏΠΎΠΊΠ° Π½Π΅ ΠΏΠ΅ΡΠ΅ΡΠ°Π³Π½Π΅ΠΌ ΡΠ΅ΡΠ΅Π· Π·Π°ΠΏΠΈΡΡ 2-Π³ΠΎ
SELECT
CASE
-- Π΅ΡΠ»ΠΈ Π½ΠΈΡΠ΅Π³ΠΎ Π½Π΅ Π½Π°ΠΉΠ΄Π΅Π½ΠΎ Π΄Π»Ρ ΠΊΠ»ΡΡΠ° 1-ΠΉ Π·Π°ΠΏΠΈΡΠΈ
WHEN X._r IS NOT DISTINCT FROM NULL THEN
T.list[2:] -- ΡΠ±ΠΈΡΠ°Π΅ΠΌ Π΅Π΅ ΠΈΠ· ΡΠΏΠΈΡΠΊΠ°
-- Π΅ΡΠ»ΠΈ ΠΌΡ ΠΠ ΠΏΠ΅ΡΠ΅ΡΠ΅ΠΊΠ»ΠΈ ΠΏΡΠΈΠΊΠ»Π°Π΄Π½ΠΎΠΉ ΠΊΠ»ΡΡ 2-ΠΉ Π·Π°ΠΏΠΈΡΠΈ
WHEN X.not_cross THEN
T.list -- ΠΏΡΠΎΡΡΠΎ ΠΏΡΠΎΡΡΠ³ΠΈΠ²Π°Π΅ΠΌ ΡΠΎΡ ΠΆΠ΅ ΡΠΏΠΈΡΠΎΠΊ Π±Π΅Π· ΠΌΠΎΠ΄ΠΈΡΠΈΠΊΠ°ΡΠΈΠΉ
-- Π΅ΡΠ»ΠΈ Π² ΡΠΏΠΈΡΠΊΠ΅ ΡΠΆΠ΅ Π½Π΅Ρ 2-ΠΉ Π·Π°ΠΏΠΈΡΠΈ
WHEN T.list[2] IS NULL THEN
-- ΠΏΡΠΎΡΡΠΎ Π²ΠΎΠ·Π²ΡΠ°ΡΠ°Π΅ΠΌ ΠΏΡΡΡΠΎΠΉ ΡΠΏΠΈΡΠΎΠΊ
'{}'
-- ΠΏΠ΅ΡΠ΅ΡΠΎΡΡΠΈΡΠΎΠ²ΡΠ²Π°Π΅ΠΌ ΡΠ»ΠΎΠ²Π°ΡΡ, ΡΠ±ΠΈΡΠ°Ρ 1-Ρ Π·Π°ΠΏΠΈΡΡ ΠΈ Π΄ΠΎΠ±Π°Π²Π»ΡΡ ΠΏΠΎΡΠ»Π΅Π΄Π½ΡΡ ΠΈΠ· Π½Π°ΠΉΠ΄Π΅Π½Π½ΡΡ
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 ( -- "ΠΌΠ°ΡΠ΅ΡΠΈΠ°Π»ΠΈΠ·ΡΠ΅ΠΌ" record
SELECT
CASE
-- Π΅ΡΠ»ΠΈ Π²ΡΠ΅-ΡΠ°ΠΊΠΈ "ΠΏΠ΅ΡΠ΅ΡΠ°Π³Π½ΡΠ»ΠΈ" ΡΠ΅ΡΠ΅Π· 2-Ρ Π·Π°ΠΏΠΈΡΡ
WHEN NOT T.not_cross
-- ΡΠΎ Π½ΡΠΆΠ½Π°Ρ Π·Π°ΠΏΠΈΡΡ - ΠΏΠ΅ΡΠ²Π°Ρ ΠΈΠ· ΡΠΏΠΏΠΈΡΠΊΠ°
THEN T.list[1]
ELSE ( -- Π΅ΡΠ»ΠΈ Π½Π΅ ΠΏΠ΅ΡΠ΅ΡΠ΅ΠΊΠ»ΠΈ, ΡΠΎ ΠΊΠ»ΡΡ ΠΎΡΡΠ°Π»ΡΡ ΠΊΠ°ΠΊ Π² ΠΏΡΠ΅Π΄ΡΠ΄ΡΡΠ΅ΠΉ Π·Π°ΠΏΠΈΡΠΈ - ΠΎΡΡΠ°Π»ΠΊΠΈΠ²Π°Π΅ΠΌΡΡ ΠΎΡ Π½Π΅Π΅
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
-- Π΅ΡΠ»ΠΈ 2-ΠΉ Π·Π°ΠΏΠΈΡΠΈ ΡΠΆΠ΅ Π½Π΅Ρ Π² ΡΠΏΠΈΡΠΊΠ΅, Π½ΠΎ ΠΌΡ Ρ
ΠΎΡΡ ΡΡΠΎ-ΡΠΎ Π½Π°ΡΠ»ΠΈ
WHEN list[2] IS NULL AND _r IS DISTINCT FROM NULL THEN
TRUE
ELSE -- Π½ΠΈΡΠ΅Π³ΠΎ Π½Π΅ Π½Π°ΡΠ»ΠΈ ΠΈΠ»ΠΈ "ΠΏΠ΅ΡΠ΅ΡΠ°Π³Π½ΡΠ»ΠΈ"
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 -- ΠΎΠ³ΡΠ°Π½ΠΈΡΠΈΠ²Π°Π΅ΠΌ ΡΡΡ ΠΊΠΎΠ»ΠΈΡΠ΅ΡΡΠ²ΠΎ
T.list IS DISTINCT FROM '{}' -- ΠΈΠ»ΠΈ ΠΏΠΎΠΊΠ° ΡΠΏΠΈΡΠΎΠΊ Π½Π΅ ΠΊΠΎΠ½ΡΠΈΠ»ΡΡ
)
-- #3 : "ΡΠ°Π·Π²ΠΎΡΠ°ΡΠΈΠ²Π°Π΅ΠΌ" Π·Π°ΠΏΠΈΡΠΈ - ΠΏΠΎΡΡΠ΄ΠΎΠΊ Π³Π°ΡΠ°Π½ΡΠΈΡΠΎΠ²Π°Π½ ΠΏΠΎ ΠΏΠΎΡΡΡΠΎΠ΅Π½ΠΈΡ
SELECT
(rv).*
FROM
T
WHERE
not_cross; -- Π±Π΅ΡΠ΅ΠΌ ΡΠΎΠ»ΡΠΊΠΎ "Π½Π΅ΠΏΠ΅ΡΠ΅ΡΠ΅ΠΊΠ°ΡΡΠΈΠ΅" Π·Π°ΠΏΠΈΡΠΈ
Sidaas darteed, waxaan ka baayacmushtarray 50% xogta la akhriyo 20% wakhtiga fulinta. Taasi waa, haddii aad haysatid sababo aad ku aaminsan tahay in wax akhrisku ay qaadan karto wakhti dheer (tusaale ahaan, xogta inta badan kuma jirto kaydka, oo waa inaad u tagtaa diskka), markaa habkan waxaad ku tiirsanaan kartaa wax ka yar akhriska. .
Si kastaba ha ahaatee, wakhtiga fulinta ayaa u soo baxay inuu ka fiicnaado "naive" doorashada koowaad. Laakin keebaa 3-daan doorasho ee aad isticmaalayso adiga ku gaar ah.
Source: www.habr.com