共计 5587 个字符,预计需要花费 14 分钟才能阅读完成。
剖析类查问即使是 TPC- H 中的简略剖析查问,PostgreSQL 的性能都要远远优于 MySQL。昆仑数据库继承并扩大了 PostgreSQL 在剖析类 SQL 查询处理方面的弱小能力,能够反对 TPC- H 和 TPC-DS 的所有查问。同时,一个昆仑数据库集群能够治理的数据规模远远大于 一个 PostgreSQL 实例。
一、测试环境
服务器配置:PostgreSQL 和 MySQL 别离部署在一台:亚马逊 m5.4xlarge(CPU 8cores 16 Threads, 内存 64G,存储 gp3, 通用型 SSD 卷 3000IOPS,125 MB/s 吞吐量)上。
软件版本:
PostgreSQL:PostgreSQL 12.7 on x86_64-koji-linux-gn
MySQL: percona 8.0.26-16
数据库参数配置:
PostgreSQL:shared_buffers 8192MB
MySQL: innodb_buffer_pool_size 8192MB
测试背景:PostgreSQL 和 MySQL 采纳默认的装置配置,只调整了内存参数,整个测试过程 PostgreSQL 没有任何优化行为。
二、测试数据
表的信息:
1 张事实表:lineorder
4 张维度表:customer,part,supplier,dates
表占用操作系统存储空间:19G
数据查问:11 条规范 SQL 查问测试语句(统计查问、多表关联、sum、简单条件、group by、order by 等组合形式)。
具体 SQL 语句:附录 1 TPC- H 测试 SQL。
三、测试后果
MySQL 错误信息:ERROR 3 (HY000): Error writing file ‘/kunlun2/data10/6010/tmp/MYfd=332′(OS errno 28-No space left on device)
剖析及总结: 通过比照同一 SQL 语句在 PostgreSQL 和 MySQL 执行打算,会发现 MySQL 的执行打算没有采纳最优的 join 秩序及并行操作,导致性能差。
譬如 Q2.1 MySQL 执行打算:
执行打算剖析: 上述执行打算首先将几个维度表做 join(dates 和 supplier 和 part), 失去的后果再与事实表 lineorder join, 因此得出了一个超级大的两头后果集,数据量达到 10 的 15 次方的数量(查问打算第五行返回的 rows),最终导致临时文件耗尽磁盘空间而未能实现查问。
优化计划: 通过在 SQL 语句中强制指定表 join 秩序:首先与 part 表 join 失去一个最小的数据子集,而后再与 supplier 和 dates join, 逐渐放大范畴,查问语句及查问打算的成果如下:
explain format=tree selectsum(lo_revenue) as lo_revenue, d_year as year, p_brand from ((lineorderstraight_join part on lo_partkey = p_partkey) straight_join supplier on lo_suppkey = s_suppkey) straight_join dates ON lo_orderdate = d_datekey where p_category ='MFGR#12' and s_region = 'AMERICA' group by year, p_brand order by year,p_brand;
在 MySQL 上指定 join 秩序,
同时开启动并行查问:set local innodb_parallel_read_threads=16
执行工夫:1 min 53.72 sec
而 PostgreSQL 的执行打算:
执行工夫:1133ms
PostgreSQL 主动采纳最优的 join 秩序,大大减少了最初排序的数据量,并且启动了 3 个并行宰割数据集,因而执行的性能比手动优化后的 MySQL 还快 10 倍。
论断:MySQL 执行 TPC- H 的测试,须要手动优化查问语句,即使如此,性能任然远远低于 PostgreSQL,而 PostgreSQL 默认的配置就能够达到绝对好的性能。
四、附录:测试 SQL 及查问打算
Q1.1
select sum(lo_revenue) as revenue from lineorder
join dates on lo_orderdate = d_datekey
where d_year = 1993 and lo_discount between 1 and 3 and lo_quantity < 25;
PostgreSQL 执行打算:
MySQL 执行打算:
Q1.2
select sum(lo_revenue) as revenue from lineorder
join dates on lo_orderdate = d_datekey
where d_yearmonthnum = 199401
and lo_discount between 4 and 6
and lo_quantity between 26 and 35;
PostgreSQL 执行打算:
MySQL 执行打算:
Q1.3
select sum(lo_revenue) as revenue from lineorder
join dates on lo_orderdate = d_datekey
where d_weeknuminyear = 6 and year(d_datekey) = 1994
and lo_discount between 5 and 7
and lo_quantity between 26 and 35;
PostgreSQL 执行打算:
MySQL 执行打算:
Q2.1
select sum(lo_revenue) as lo_revenue, d_year as year, p_brand
from lineorder
join dates on lo_orderdate = d_datekey
join part on lo_partkey = p_partkey
join supplier on lo_suppkey = s_suppkey
where p_category = 'MFGR#12' and s_region = 'AMERICA'
group by year, p_brand
order by year, p_brand;
PostgreSQL 执行打算:
MySQL 执行打算:
Q2.2
select sum(lo_revenue) as lo_revenue, year(d_datekey) as year, p_brand
from lineorder
join dates on lo_orderdate = d_datekey
join part on lo_partkey = p_partkey
join supplier on lo_suppkey = s_suppkey
where p_brand between 'MFGR#2221' and 'MFGR#2228' and s_region = 'ASIA'
group by year, p_brand
order by year, p_brand;
PostgreSQL 执行打算:
MySQL 执行打算:
Q2.3
select sum(lo_revenue) as lo_revenue, year(d_datekey) as year, p_brand
from lineorder
join dates on lo_orderdate = d_datekey
join part on lo_partkey = p_partkey
join supplier on lo_suppkey = s_suppkey
where p_brand = 'MFGR#2239' and s_region = 'EUROPE'
group by year, p_brand
order by year, p_brand;
PostgreSQL 执行打算:
MySQL 执行打算:
Q3.1
select c_nation, s_nation, year(d_datekey) as year, sum(lo_revenue) as lo_revenue
from lineorder
join dates on lo_orderdate = d_datekey
join customer on lo_custkey = c_custkey
join supplier on lo_suppkey = s_suppkey
where c_region = 'ASIA' and s_region = 'ASIA' and year(d_datekey) between 1992 and 1997
group by c_nation, s_nation, year
order by year asc, lo_revenue desc;
PostgreSQL 执行打算:
MySQL 执行打算:
Q3.2
select c_city, s_city, year(d_datekey) as year, sum(lo_revenue) as lo_revenue
from lineorder
join dates on lo_orderdate = d_datekey
join customer on lo_custkey = c_custkey
join supplier on lo_suppkey = s_suppkey
where c_nation = 'UNITED STATES' and s_nation = 'UNITED STATES'
and year(d_datekey) between 1992 and 1997
group by c_city, s_city, year
order by year asc, lo_revenue desc;
PostgreSQL 执行打算:
MySQL 执行打算:
Q3.3
select c_city, s_city, year(d_datekey) as year, sum(lo_revenue) as lo_revenue
from lineorder
join dates on lo_orderdate = d_datekey
join customer on lo_custkey = c_custkey
join supplier on lo_suppkey = s_suppkey
where (c_city='UNITED KI1' or c_city='UNITED KI5')
and (s_city='UNITED KI1' or s_city='UNITED KI5')
and year(d_datekey) between 1992 and 1997
group by c_city, s_city, year
order by year asc, lo_revenue desc;
PostgreSQL 执行打算:
MySQL 执行打算:
Q4.1
select year(d_datekey) as year, c_nation, sum(lo_revenue) - sum(lo_supplycost) as profit
from lineorder
join dates on lo_orderdate = d_datekey
join customer on lo_custkey = c_custkey
join supplier on lo_suppkey = s_suppkey
join part on lo_partkey = p_partkey
where c_region = 'AMERICA' and s_region = 'AMERICA' and (p_mfgr = 'MFGR#1' or p_mfgr = 'MFGR#2')
group by year, c_nation
order by year, c_nation;
PostgreSQL 执行打算:
MySQL 执行打算:
Q4.2
select year(d_datekey) as year, s_nation, p_category, sum(lo_revenue) - sum(lo_supplycost) as profit
from lineorder
join dates on lo_orderdate = d_datekey
join customer on lo_custkey = c_custkey
join supplier on lo_suppkey = s_suppkey
join part on lo_partkey = p_partkey
where c_region = 'AMERICA'and s_region = 'AMERICA'
and (year(d_datekey) = 1997 or year(d_datekey) = 1998)
and (p_mfgr = 'MFGR#1' or p_mfgr = 'MFGR#2')
group by year, s_nation, p_category
order by year, s_nation, p_category;
PostgreSQL 执行打算:
MySQL 执行打算:
END
昆仑数据库是一个 HTAP NewSQL 分布式数据库管理系统,能够满足用户对海量关系数据的存储管理和利用的全方位需要。
利用开发者和 DBA 的应用昆仑数据库的体验与单机 MySQL 和单机 PostgreSQL 简直完全相同,因为首先昆仑数据库反对 PostgreSQL 和 MySQL 双协定,反对规范 SQL:2011 的 DML 语法和性能以及 PostgreSQL 和 MySQL 对规范 SQL 的扩大。同时,昆仑数据库集群反对程度弹性扩容,数据主动拆分,分布式事务处理和分布式查询处理,强壮的容错容灾能力,欠缺直观的监测剖析告警能力,集群数据备份和复原等 罕用的 DBA 数据管理和操作。所有这些性能无需任何利用零碎侧的编码工作,也无需 DBA 人工染指,不停服不影响业务失常运行。
昆仑数据库具备全面的 OLAP 数据分析能力,通过了 TPC- H 和 TPC-DS 规范测试集,能够实时剖析最新的业务数据,帮忙用户发掘出数据的价值。昆仑数据库反对私有云和公有云环境的部署,能够与 docker,k8s 等云基础设施无缝合作,能够轻松搭建云数据库服务。
请拜访 http://www.zettadb.com/ 获取更多信息并且下载昆仑数据库软件、文档和材料。
KunlunDB 我的项目已开源
【GitHub:】
https://github.com/zettadb
【Gitee:】
https://gitee.com/zettadb