MaCPU emazuva ano ane akawanda macores. Kwemakore, zvikumbiro zvanga zvichitumira mibvunzo kune dhatabhesi zvakafanana. Kana iri mubvunzo wemushumo pamitsetse yakawanda mutafura, inomhanya nekukurumidza kana uchishandisa akawanda maCPU, uye PostgreSQL yakakwanisa kuita izvi kubvira vhezheni 9.6.
Zvakatora makore matatu kuti tishandise iyo yakafanana query ficha - taifanira kunyorazve kodhi pamatanho akasiyana ekuita mubvunzo. PostgreSQL 3 yakaunza zvivakwa kuti iwedzere kuvandudza kodhi. Mushanduro dzinotevera, mamwe marudzi emibvunzo anoitwa achitevedzana.
Zvibvumirano
- Usagone kuita parallel execution kana ese macores akatobatikana, zvikasadaro zvimwe zvikumbiro zvinodzikira.
- Kunyanya kukosha, kuenderana kugadzirisa neakakwira WORK_MEM kukosha kunoshandisa yakawanda ndangariro - yega hashi kujoinha kana rudzi kunotora work_mem ndangariro.
- Yakaderera latency OLTP mibvunzo haigone kukwidziridzwa neparallel execution. Uye kana mubvunzo ukadzosa mutsara mumwe, parallel processing inongodzinotsa.
- Vagadziri vanofarira kushandisa iyo TPC-H bhenji. Pamwe iwe une mibvunzo yakafanana yekunyatso enzanirana execution.
- Chete SARUDZA mibvunzo isina predicate kukiya inoitwa mukuwirirana.
- Dzimwe nguva indexing chaiyo iri nani pane inoteedzana tafura scanning mune parallel mode.
- Kumbomira mivhunzo uye macursor haatsigirwe.
- Window mabasa uye akaodha set aggregate mabasa haana kufambirana.
- Iwe hauwane chero chinhu muI / O basa rekuita.
- Iko hakuna kufanana kwekuronga algorithms. Asi mibvunzo ine marudzi inogona kuitwa mukuwirirana mune zvimwe zvinhu.
- Tsiva CTE (NE ...) neinested SELECT kuti igone kuenderana kugadzirisa.
- Yechitatu-bato data wrappers haisati yatsigira kuenderana kugadzirisa (asi ivo vanogona!)
- FULL OUTER JOIN haitsigirwe.
- max_rows anodzima parallel processing.
- Kana muvhunzo uine basa risina kunyorwa PARALLEL SAFE, rinenge riine thread imwechete.
- Iyo SERIALIZABLE transaction yekuzviparadzanisa level inodzima parallel process.
Test environment
Vagadziri vePostgreSQL vakaedza kudzikisa nguva yekupindura yeTPC-H benchmark mibvunzo. Dhawunirodha bhenji uye . Uku ndiko kushandiswa kusiri pamutemo kweiyo TPC-H bhenji - kwete yedhatabhesi kana hardware kuenzanisa.
- Dhaunirodha TPC-H_Tools_v2.17.3.zip (kana shanduro itsva) .
- Rename makefile.suite kuMakefile uye shandura sezvinotsanangurwa pano: . Nyora kodhi ne make command.
- Gadzira data:
./dbgen -s 10inogadzira 23 GB database. Izvi zvakakwana kuti uone mutsauko mukuita kwemibvunzo yakafanana uye isiri-yakafanana. - Shandura mafaira
tblΠ²csv Ρ forΠΈsed. - Clone iyo repository
pg_tpchuye kopira mafairacsvΠ²pg_tpch/dss/data. - Gadzira mibvunzo nemirairo
qgen. - Rodha data mudhatabhesi nemurairo
./tpch.sh.
Parallel sequential scanning
Inogona kukurumidza kwete nekuda kwekuverenga kwakafanana, asi nekuti iyo data yakapararira kune akawanda CPU cores. Mumazuva ano anoshanda masisitimu, PostgreSQL data mafaera akachengetwa zvakanaka. Nekuverenga kumberi, zvinokwanisika kuwana yakakura block kubva mukuchengetedza pane iyo PG daemon zvikumbiro. Naizvozvo, kuita kwemubvunzo hakuganhurwe ne diski I/O. Iyo inoshandisa CPU kutenderera ku:
- verenga mitsara imwe neimwe kubva pamapeji etafura;
- enzanisa tambo tsika uye mamiriro
WHERE.
Ngatimhanyei mubvunzo uri nyore select:
tpch=# explain analyze select l_quantity as sum_qty from lineitem where l_shipdate <= date '1998-12-01' - interval '105' day;
QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------
Seq Scan on lineitem (cost=0.00..1964772.00 rows=58856235 width=5) (actual time=0.014..16951.669 rows=58839715 loops=1)
Filter: (l_shipdate <= '1998-08-18 00:00:00'::timestamp without time zone)
Rows Removed by Filter: 1146337
Planning Time: 0.203 ms
Execution Time: 19035.100 msIyo sequential scan inoburitsa mitsara yakawandisa isina kuunganidzwa, saka mubvunzo unoitwa neiyo imwechete CPU musimboti.
Kana iwe ukawedzera SUM(), unogona kuona kuti mafambiro maviri ebasa achabatsira kukurumidza kubvunza:
explain analyze select sum(l_quantity) as sum_qty from lineitem where l_shipdate <= date '1998-12-01' - interval '105' day;
QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------------------------
Finalize Aggregate (cost=1589702.14..1589702.15 rows=1 width=32) (actual time=8553.365..8553.365 rows=1 loops=1)
-> Gather (cost=1589701.91..1589702.12 rows=2 width=32) (actual time=8553.241..8555.067 rows=3 loops=1)
Workers Planned: 2
Workers Launched: 2
-> Partial Aggregate (cost=1588701.91..1588701.92 rows=1 width=32) (actual time=8547.546..8547.546 rows=1 loops=3)
-> Parallel Seq Scan on lineitem (cost=0.00..1527393.33 rows=24523431 width=5) (actual time=0.038..5998.417 rows=19613238 loops=3)
Filter: (l_shipdate <= '1998-08-18 00:00:00'::timestamp without time zone)
Rows Removed by Filter: 382112
Planning Time: 0.241 ms
Execution Time: 8555.131 msParallel aggregation
Iyo Parallel Seq Scan node inogadzira mitsetse yekuunganidza zvishoma. Iyo "Partial Aggregate" node inocheka mitsetse iyi uchishandisa SUM(). Pakupedzisira, iyo SUM counter kubva kune yega yega yevashandi maitiro inounganidzwa ne "Gather" node.
Mhedzisiro yekupedzisira inoverengerwa ne "Finalize Aggregate" node. Kana iwe uine ako ega ekuunganidza mabasa, usakanganwa kumaka se "parallel safe".
Nhamba yemaitiro evashandi
Huwandu hwemaitiro evashandi hunogona kuwedzerwa pasina kutangazve server:
explain analyze select sum(l_quantity) as sum_qty from lineitem where l_shipdate <= date '1998-12-01' - interval '105' day;
QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------------------------
Finalize Aggregate (cost=1589702.14..1589702.15 rows=1 width=32) (actual time=8553.365..8553.365 rows=1 loops=1)
-> Gather (cost=1589701.91..1589702.12 rows=2 width=32) (actual time=8553.241..8555.067 rows=3 loops=1)
Workers Planned: 2
Workers Launched: 2
-> Partial Aggregate (cost=1588701.91..1588701.92 rows=1 width=32) (actual time=8547.546..8547.546 rows=1 loops=3)
-> Parallel Seq Scan on lineitem (cost=0.00..1527393.33 rows=24523431 width=5) (actual time=0.038..5998.417 rows=19613238 loops=3)
Filter: (l_shipdate <= '1998-08-18 00:00:00'::timestamp without time zone)
Rows Removed by Filter: 382112
Planning Time: 0.241 ms
Execution Time: 8555.131 msChii chiri kuitika pano? Paiva ne2 nguva dzakawanda dzekushanda, uye chikumbiro chakava chete 1,6599 nguva nokukurumidza. Masvomhu anonakidza. Takanga tine maitiro maviri evashandi uye mutungamiri mumwe. Mushure mekuchinja kwakava 2 + 1.
Yedu yepamusoro yekumhanyisa kubva kunoenderana kugadzirisa: 5/3 = 1,66 (6) nguva.
Sei kushanda?
Maitiro acho
Kuita chikumbiro nguva dzose kunotanga neinotungamira maitiro. Mutungamiri anoita zvese zvisingaenderani uye zvimwe zvakafanana kugadzirisa. Mamwe maitiro anoita zvikumbiro zvakafanana anonzi maitiro evashandi. Parallel processing inoshandisa zvivakwa (kubva mushanduro 9.4). Sezvo zvimwe zvikamu zvePostgreSQL zvichishandisa maitiro kwete tambo, mubvunzo une 3 mushandi maitiro anogona kunge ari 4 nguva nekukurumidza kupfuura echinyakare kugadzirisa.
Kubatana
Maitiro evashandi anotaurirana nemutungamiri kuburikidza nemutsara wemeseji (zvichienderana nendangariro dzakagovaniswa). Maitiro ega ega ane 2 mitsetse: yezvikanganiso uye yematuples.
Ndeapi mafambiro ebasa anodiwa?
Muganho wepasi unotsanangurwa neparameter . Mumhanyi wekukumbira anobva atora maitiro evashandi kubva padziva akaganhurirwa neparameter . Mhedzisiro yekupedzisira ndeye , kureva, nhamba yose yemashure maitiro.
Kana zvaisaita kugovera maitiro evashandi, kugadzirisa kuchave kumwe-maitiro.
Kuronga kwemubvunzo kunogona kuderedza mafambiro ebasa zvichienderana nehukuru hwetafura kana index. Pane zvimiro zveizvi ΠΈ .
set min_parallel_table_scan_size='8MB'
8MB table => 1 worker
24MB table => 2 workers
72MB table => 3 workers
x => log(x / min_parallel_table_scan_size) / log(3) + 1 workerNguva dzose tafura iri 3 nguva yakakura kudarika min_parallel_(index|table)_scan_size, Postgres inowedzera maitiro evashandi. Huwandu hwekufamba kwebasa haubvi pamitengo. Kutsamira kwedenderedzwa kunoita kuti kuita kwakaoma kuome. Pane kudaro, murongi anoshandisa mitemo iri nyore.
Mukuita, iyi mitemo haisi nguva dzose yakakodzera kugadzirwa, saka unogona kuchinja nhamba yevashandi maitiro kune imwe tafura: ALTER TABLE ... SET (parallel_workers = N).
Sei parallel processing isiri kushandiswa?
Pamusoro peiyo refu runyorwa rwezvirambidzo, kune zvakare mutengo wekutarisa:
- kudzivirira kuenderana kwekugadziriswa kwezvikumbiro zvipfupi. Iyi parameter inofungidzira nguva yekugadzirira ndangariro, kutanga maitiro, uye yekutanga data exchange.
: kutaurirana pakati pemutungamiri nevashandi kunogona kunonoka maererano nehuwandu hwema tuples kubva kumabasa ekushanda. Iyi parameter inoverenga mutengo wekuchinjana data.
Nested Loop Joins
PostgreSQL 9.6+ ΠΌΠΎΠΆΠ΅Ρ Π²ΡΠΏΠΎΠ»Π½ΡΡΡ Π²Π»ΠΎΠΆΠ΅Π½Π½ΡΠ΅ ΡΠΈΠΊΠ»Ρ ΠΏΠ°ΡΠ°Π»Π»Π΅Π»ΡΠ½ΠΎ β ΡΡΠΎ ΠΏΡΠΎΡΡΠ°Ρ ΠΎΠΏΠ΅ΡΠ°ΡΠΈΡ.
explain (costs off) select c_custkey, count(o_orderkey)
from customer left outer join orders on
c_custkey = o_custkey and o_comment not like '%special%deposits%'
group by c_custkey;
QUERY PLAN
--------------------------------------------------------------------------------------
Finalize GroupAggregate
Group Key: customer.c_custkey
-> Gather Merge
Workers Planned: 4
-> Partial GroupAggregate
Group Key: customer.c_custkey
-> Nested Loop Left Join
-> Parallel Index Only Scan using customer_pkey on customer
-> Index Scan using idx_orders_custkey on orders
Index Cond: (customer.c_custkey = o_custkey)
Filter: ((o_comment)::text !~~ '%special%deposits%'::text)Iko kuunganidzwa kunoitika padanho rekupedzisira, saka Nested Loop Kuruboshwe Kujoinha ibasa rakafanana. Parallel Index Chete Scan yakaunzwa mushanduro 10 chete. Inoshanda zvakafanana neparallel serial scanning. Condition c_custkey = o_custkey inoverenga odha imwe patambo yemutengi. Saka hazvina kufambirana.
Hash Join
Nzira yega yega yevashandi inogadzira tafura yayo yehashi kusvika PostgreSQL 11. Uye kana pane zvinopfuura zvina zvezvirongwa izvi, kushanda hakuzogadziriswi. Mushanduro itsva, tafura yehashi inogoverwa. Maitiro ega ega evashandi anogona kushandisa WORK_MEM kugadzira tafura yehashi.
select
l_shipmode,
sum(case
when o_orderpriority = '1-URGENT'
or o_orderpriority = '2-HIGH'
then 1
else 0
end) as high_line_count,
sum(case
when o_orderpriority <> '1-URGENT'
and o_orderpriority <> '2-HIGH'
then 1
else 0
end) as low_line_count
from
orders,
lineitem
where
o_orderkey = l_orderkey
and l_shipmode in ('MAIL', 'AIR')
and l_commitdate < l_receiptdate
and l_shipdate < l_commitdate
and l_receiptdate >= date '1996-01-01'
and l_receiptdate < date '1996-01-01' + interval '1' year
group by
l_shipmode
order by
l_shipmode
LIMIT 1;
QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Limit (cost=1964755.66..1964961.44 rows=1 width=27) (actual time=7579.592..7922.997 rows=1 loops=1)
-> Finalize GroupAggregate (cost=1964755.66..1966196.11 rows=7 width=27) (actual time=7579.590..7579.591 rows=1 loops=1)
Group Key: lineitem.l_shipmode
-> Gather Merge (cost=1964755.66..1966195.83 rows=28 width=27) (actual time=7559.593..7922.319 rows=6 loops=1)
Workers Planned: 4
Workers Launched: 4
-> Partial GroupAggregate (cost=1963755.61..1965192.44 rows=7 width=27) (actual time=7548.103..7564.592 rows=2 loops=5)
Group Key: lineitem.l_shipmode
-> Sort (cost=1963755.61..1963935.20 rows=71838 width=27) (actual time=7530.280..7539.688 rows=62519 loops=5)
Sort Key: lineitem.l_shipmode
Sort Method: external merge Disk: 2304kB
Worker 0: Sort Method: external merge Disk: 2064kB
Worker 1: Sort Method: external merge Disk: 2384kB
Worker 2: Sort Method: external merge Disk: 2264kB
Worker 3: Sort Method: external merge Disk: 2336kB
-> Parallel Hash Join (cost=382571.01..1957960.99 rows=71838 width=27) (actual time=7036.917..7499.692 rows=62519 loops=5)
Hash Cond: (lineitem.l_orderkey = orders.o_orderkey)
-> Parallel Seq Scan on lineitem (cost=0.00..1552386.40 rows=71838 width=19) (actual time=0.583..4901.063 rows=62519 loops=5)
Filter: ((l_shipmode = ANY ('{MAIL,AIR}'::bpchar[])) AND (l_commitdate < l_receiptdate) AND (l_shipdate < l_commitdate) AND (l_receiptdate >= '1996-01-01'::date) AND (l_receiptdate < '1997-01-01 00:00:00'::timestamp without time zone))
Rows Removed by Filter: 11934691
-> Parallel Hash (cost=313722.45..313722.45 rows=3750045 width=20) (actual time=2011.518..2011.518 rows=3000000 loops=5)
Buckets: 65536 Batches: 256 Memory Usage: 3840kB
-> Parallel Seq Scan on orders (cost=0.00..313722.45 rows=3750045 width=20) (actual time=0.029..995.948 rows=3000000 loops=5)
Planning Time: 0.977 ms
Execution Time: 7923.770 msMubvunzo 12 kubva kuTPC-H unoratidza zvakajeka kubatana kwehashi. Maitiro ega ega evashandi anobatsira mukugadzira tafura yehashi.
Merge Join
Kubatanidza kujoinwa hakuenderane mune zvakasikwa. Usanetseke kana iri iri danho rekupedzisira remubvunzo - rinogona kuramba richimhanya zvakafanana.
-- Query 2 from TPC-H
explain (costs off) select s_acctbal, s_name, n_name, p_partkey, p_mfgr, s_address, s_phone, s_comment
from part, supplier, partsupp, nation, region
where
p_partkey = ps_partkey
and s_suppkey = ps_suppkey
and p_size = 36
and p_type like '%BRASS'
and s_nationkey = n_nationkey
and n_regionkey = r_regionkey
and r_name = 'AMERICA'
and ps_supplycost = (
select
min(ps_supplycost)
from partsupp, supplier, nation, region
where
p_partkey = ps_partkey
and s_suppkey = ps_suppkey
and s_nationkey = n_nationkey
and n_regionkey = r_regionkey
and r_name = 'AMERICA'
)
order by s_acctbal desc, n_name, s_name, p_partkey
LIMIT 100;
QUERY PLAN
----------------------------------------------------------------------------------------------------------
Limit
-> Sort
Sort Key: supplier.s_acctbal DESC, nation.n_name, supplier.s_name, part.p_partkey
-> Merge Join
Merge Cond: (part.p_partkey = partsupp.ps_partkey)
Join Filter: (partsupp.ps_supplycost = (SubPlan 1))
-> Gather Merge
Workers Planned: 4
-> Parallel Index Scan using <strong>part_pkey</strong> on part
Filter: (((p_type)::text ~~ '%BRASS'::text) AND (p_size = 36))
-> Materialize
-> Sort
Sort Key: partsupp.ps_partkey
-> Nested Loop
-> Nested Loop
Join Filter: (nation.n_regionkey = region.r_regionkey)
-> Seq Scan on region
Filter: (r_name = 'AMERICA'::bpchar)
-> Hash Join
Hash Cond: (supplier.s_nationkey = nation.n_nationkey)
-> Seq Scan on supplier
-> Hash
-> Seq Scan on nation
-> Index Scan using idx_partsupp_suppkey on partsupp
Index Cond: (ps_suppkey = supplier.s_suppkey)
SubPlan 1
-> Aggregate
-> Nested Loop
Join Filter: (nation_1.n_regionkey = region_1.r_regionkey)
-> Seq Scan on region region_1
Filter: (r_name = 'AMERICA'::bpchar)
-> Nested Loop
-> Nested Loop
-> Index Scan using idx_partsupp_partkey on partsupp partsupp_1
Index Cond: (part.p_partkey = ps_partkey)
-> Index Scan using supplier_pkey on supplier supplier_1
Index Cond: (s_suppkey = partsupp_1.ps_suppkey)
-> Index Scan using nation_pkey on nation nation_1
Index Cond: (n_nationkey = supplier_1.s_nationkey)Iyo "Merge Join" node iri pamusoro pe "Gather Merge". Saka kubatanidza hakushandisi parallel processing. Asi iyo "Parallel Index Scan" node ichiri kubatsira nechikamu part_pkey.
Kubatanidza nezvikamu
MuPostgreSQL 11 yakaremara nekusarudzika: ine inodhura kwazvo kuronga. Matafura ane kupatsanurwa kwakafanana anogona kubatanidzwa kupatsanura nekupatsanura. Nenzira iyi Postgres ichashandisa madiki hashi matafura. Kubatana kwega kwega kwezvikamu kunogona kuenderana.
tpch=# set enable_partitionwise_join=t;
tpch=# explain (costs off) select * from prt1 t1, prt2 t2
where t1.a = t2.b and t1.b = 0 and t2.b between 0 and 10000;
QUERY PLAN
---------------------------------------------------
Append
-> Hash Join
Hash Cond: (t2.b = t1.a)
-> Seq Scan on prt2_p1 t2
Filter: ((b >= 0) AND (b <= 10000))
-> Hash
-> Seq Scan on prt1_p1 t1
Filter: (b = 0)
-> Hash Join
Hash Cond: (t2_1.b = t1_1.a)
-> Seq Scan on prt2_p2 t2_1
Filter: ((b >= 0) AND (b <= 10000))
-> Hash
-> Seq Scan on prt1_p2 t1_1
Filter: (b = 0)
tpch=# set parallel_setup_cost = 1;
tpch=# set parallel_tuple_cost = 0.01;
tpch=# explain (costs off) select * from prt1 t1, prt2 t2
where t1.a = t2.b and t1.b = 0 and t2.b between 0 and 10000;
QUERY PLAN
-----------------------------------------------------------
Gather
Workers Planned: 4
-> Parallel Append
-> Parallel Hash Join
Hash Cond: (t2_1.b = t1_1.a)
-> Parallel Seq Scan on prt2_p2 t2_1
Filter: ((b >= 0) AND (b <= 10000))
-> Parallel Hash
-> Parallel Seq Scan on prt1_p2 t1_1
Filter: (b = 0)
-> Parallel Hash Join
Hash Cond: (t2.b = t1.a)
-> Parallel Seq Scan on prt2_p1 t2
Filter: ((b >= 0) AND (b <= 10000))
-> Parallel Hash
-> Parallel Seq Scan on prt1_p1 t1
Filter: (b = 0)Chinhu chikuru ndechokuti kubatana muzvikamu kunofanana chete kana zvikamu izvi zvakakura zvakakwana.
Parallel Append
inogona kushandiswa pachinzvimbo chezvivharo zvakasiyana mukufambiswa kwebasa kwakasiyana. Izvi zvinowanzoitika neUNION YESE mibvunzo. Izvo zvisingabatsiri ndezvishoma parallelism, nekuti yega yega yevashandi maitiro anongogadzirisa 1 chikumbiro.
Kune maviri maitiro evashandi ari kushanda pano, kunyangwe mana akagoneswa.
tpch=# explain (costs off) select sum(l_quantity) as sum_qty from lineitem where l_shipdate <= date '1998-12-01' - interval '105' day union all select sum(l_quantity) as sum_qty from lineitem where l_shipdate <= date '2000-12-01' - interval '105' day;
QUERY PLAN
------------------------------------------------------------------------------------------------
Gather
Workers Planned: 2
-> Parallel Append
-> Aggregate
-> Seq Scan on lineitem
Filter: (l_shipdate <= '2000-08-18 00:00:00'::timestamp without time zone)
-> Aggregate
-> Seq Scan on lineitem lineitem_1
Filter: (l_shipdate <= '1998-08-18 00:00:00'::timestamp without time zone)Izvo zvakakosha zvakasiyana
- WORK_MEM inodzikamisa ndangariro pamaitiro, kwete mibvunzo chete: work_mem maitiro kubatana = kurangarira kwakawanda.
- - vangani vashandi vanogadzirisa chirongwa chekushandisa chinozoshandisa kuenderana kugadzirisa kubva kuhurongwa.
- - inogadzirisa huwandu hwese hwevashandi maitiro kune huwandu hweCPU cores pane server.
- - zvakafanana, asi nokuda kwemaitiro ekushanda akafanana.
Migumisiro
Nezve vhezheni 9.6, parallel process inogona kuvandudza zvakanyanya mashandiro emibvunzo yakaoma inotarisa mitsetse yakawanda kana indexes. MuPostgreSQL 10, parallel process inogoneswa nekusarudzika. Rangarira kuidzima pamaseva ane hombe yeOLTP yebasa. Sequential scans kana index scans inopedza zvakawanda zviwanikwa. Kana usiri kumhanyisa rondedzero pane yese dataset, unogona kuvandudza mashandiro emubvunzo nekungowedzera zvisipo kana kushandisa kupatsanura kwakaringana.
nezvakanyorwa
Source: www.habr.com