(转)explain、db2exfmt 命令的使用:文本输出执行计划
原文:http://blog.51cto.com/freebile/1068610
db2有图形执行计划显示工具,如果没有图形环境,如unix主机,可以生成文本的
文件来显示执行计划
1.如果第一次执行,请先 connect to dbname,
执行db2 -tvf $HOME/sqllib/misc/EXPLAIN.DDL建立执行计划表
2.db2 set current explain mode explain
设置成解释模式,并不真正执行下面将发出的sql命令
3.db2 "select count(*) from DOA.ODS_SFC_SALE_BARCODE_T"
执行你想要分析的sql语句
4.db2 set current explain mode no
取消解释模式
5.db2exfmt -d sample -g TIC -w -l -s % -n % -o db2exmt.out
执行计划输出到文件db2exmt.out
相关参数设置可以参考
(This application formats the contents of the Explain tables.
Given a database name and other qualifying information, this tool will,
query the Explain tables for information and format the results.
Syntax is:
db2exfmt [[-1] [-d <dbname>;] [-e <schema>;] [-f O] [-h] [-l] [-n <name>;] [-o <outfile>;]
[-s <schema>;] [-t]] [-u <user>; <pw>;] [-w <timestamp>;] [-# <sectnbr>;] [-v <srcvers>;]
Input Fields:
-d <dbname>; = database name containing packages
-e <schema>; = Explain table schema
-f O = Formatting flags (O = Operator summary)
-g[x] [O[T|F]IC] - Graph plan. x - turn off options (default is to turn them on) Options include:
O = only generate graph
T = Include Total Cost in graph
F = Include First Tuple Cost in graph
I = Include I/O Cost in graph
C = Include Cardinality in graph
Any combination of these options is allowed,
except 'F' and 'T', which are mutually exclusive.
-h = help
-l = respect package name case
-n <name>; = name of source of Explain
request (SOURCE_NAME)
-no_map_char = do no map a non-printable character to a '.'
-no_prompt = do not prompt for user input
-o <outfile>; = name of output file
-r <requester>; = id of explain requester
-s <schema>; = Schema or qualifier of source of Explain
request (SOURCE_SCHEMA)
-t = terminal output desired
-u <user>; <pw>; = user ID and password for connecting to database
-v <srcvers>; = Source Version of source of Explain request (default %)
-w <timestamp>; = Explain timestamp (use -1 to get newest Explain request)
-# <sectnbr>; = section number in source (use zero for all sections)
-1 = Use defaults -e % -n % -s % -v % -w -1 -# 0 If Explain schema is not supplied, the contents of the environment
variable $USER, or $USERNAME will be used as a default. If this
variable is not found, the user will be prompted for an Explain schema.
Source name, source schema, and Explain timestamp may be supplied
in LIKE predicate form, which allows percent sign (%) and
underscore (_) to be used as pattern matching characters to select
multiple sources with one invocation.
Prompting will occur for all fields that are not supplied or are
incompletely specified (except for the -h, -l and -no_map_char options).
If -o is specified without a file name, and -t is not specified,
the user will be prompted for a file name (the default name is
db2exfmt.out).
If neither -o nor -t is specified, the user will be prompted for
a file name (the default is terminal output).
If -o and -t are both specified, then the output will be directed
to the terminal.
)
6.查看输出文件分析sql的运行开销,示例输出如下
-------------------
DB2 Universal Database Version 9.5, 5622-044 (c) Copyright IBM Corp. 1991, 2007
Licensed Material - Program Property of IBM
IBM DATABASE 2 Explain Table Format Tool
******************** EXPLAIN INSTANCE ********************
DB2_VERSION: 09.05.4
SOURCE_NAME: SQLC2G15
SOURCE_SCHEMA: NULLID
SOURCE_VERSION:
EXPLAIN_TIME: 2012-11-23-16.02.53.924371
EXPLAIN_REQUESTER: DOAADMIN
Database Context:
----------------
Parallelism: None
CPU Speed: 4.959615e-07
Comm Speed: 100
Buffer Pool size: 438463
Sort Heap size: 256
Database Heap size: 2645
Lock List size: 100
Maximum Lock List: 20
Average Applications: 1
Locks Available: 1280
Package Context:
---------------
SQL Type: Dynamic
Optimization Level: 5
Blocking: Block All Cursors
Isolation Level: Cursor Stability
---------------- STATEMENT 1 SECTION 201 ----------------
QUERYNO: 3
QUERYTAG: CLP
Statement Type: Select
Updatable: No
Deletable: No
Query Degree: 1
Original Statement:
------------------
select count(*)
from DOA.ODS_SFC_SALE_BARCODE_T
Optimized Statement:
-------------------
SELECT Q3.$C0
FROM
(SELECT COUNT(*)
FROM
(SELECT $RID$
FROM DOA.ODS_SFC_SALE_BARCODE_T AS Q1) AS Q2) AS Q3
Access Plan:
-----------
Total Cost: 51435.9
Query Degree: 1
Rows
RETURN
( 1)
Cost
I/O
|
1
GRPBY
( 2)
51435.9
2027
|
765027
IXSCAN
( 3)
51341.1
2027
|
765027
INDEX: DOA
S_SALE_T1
Q1
Extended Diagnostic Information:
--------------------------------
No extended Diagnostic Information for this statement.
Plan Details:
-------------
1) RETURN: (Return Result)
Cumulative Total Cost: 51435.9
Cumulative CPU Cost: 1.53426e+09
Cumulative I/O Cost: 2027
Cumulative Re-Total Cost: 747.847
Cumulative Re-CPU Cost: 1.50787e+09
Cumulative Re-I/O Cost: 0
Cumulative First Row Cost: 51435.9
Estimated Bufferpool Buffers: 2028
Arguments:
---------
BLDLEVEL: (Build level)
DB2 v9.5.0.4 : s090429
HEAPUSE : (Maximum Statement Heap Usage)
80 Pages
PREPTIME: (Statement prepare time)
1 milliseconds
STMTHEAP: (Statement heap size)
6402
Input Streams:
-------------
3) From Operator #2
Estimated number of rows: 1
Number of columns: 1
Subquery predicate ID: Not Applicable
Column Names:
------------
+Q4.$C0
2) GRPBY : (Group By)
Cumulative Total Cost: 51435.9
Cumulative CPU Cost: 1.53426e+09
Cumulative I/O Cost: 2027
Cumulative Re-Total Cost: 747.847
Cumulative Re-CPU Cost: 1.50787e+09
Cumulative Re-I/O Cost: 0
Cumulative First Row Cost: 51435.9
Estimated Bufferpool Buffers: 2028
Arguments:
---------
AGGMODE : (Aggregration Mode)
COMPLETE
GROUPBYC: (Group By columns)
FALSE
GROUPBYN: (Number of Group By columns)
0
ONEFETCH: (One Fetch flag)
FALSE
Input Streams:
-------------
2) From Operator #3
Estimated number of rows: 765027
Number of columns: 0
Subquery predicate ID: Not Applicable
Output Streams:
--------------
3) To Operator #1
Estimated number of rows: 1
Number of columns: 1
Subquery predicate ID: Not Applicable
Column Names:
------------
+Q4.$C0
3) IXSCAN: (Index Scan)
Cumulative Total Cost: 51341.1
Cumulative CPU Cost: 1.343e+09
Cumulative I/O Cost: 2027
Cumulative Re-Total Cost: 652.991
Cumulative Re-CPU Cost: 1.31662e+09
Cumulative Re-I/O Cost: 0
Cumulative First Row Cost: 50.0327
Estimated Bufferpool Buffers: 2028
Arguments:
---------
MAXPAGES: (Maximum pages for prefetch)
2026
PREFETCH: (Type of Prefetch)
NONE
ROWLOCK : (Row Lock intent)
NEXT KEY SHARE
SCANDIR : (Scan Direction)
FORWARD
TABLOCK : (Table Lock intent)
INTENT SHARE
TBISOLVL: (Table access Isolation Level)
CURSOR STABILITY
Input Streams:
-------------
1) From Object DOA.S_SALE_T1
Estimated number of rows: 765027
Number of columns: 1
Subquery predicate ID: Not Applicable
Column Names:
------------
+Q1.$RID$
Output Streams:
--------------
2) To Operator #2
Estimated number of rows: 765027
Number of columns: 0
Subquery predicate ID: Not Applicable
Objects Used in Access Plan:
---------------------------
Schema: DOA
Name: ODS_SFC_SALE_BARCODE_T
Type: Table (reference only)
Schema: DOA
Name: S_SALE_T1
Type: Index
Time of creation: 2011-01-15-14.03.41.217506
Last statistics update: 2012-04-23-18.45.51.142593
Number of columns: 1
Number of rows: 765027
Width of rows: -1
Number of buffer pool pages: 50996
Distinct row values: No
Tablespace name: TS_INDEX_4K
Tablespace overhead: 24.100000
Tablespace transfer rate: 0.900000
Source for statistics: Single Node
Prefetch page count: 128
Container extent page count: 32
Index clustering statistic: 97.000000
Index leaf pages: 2026
Index tree levels: 3
Index full key cardinality: 18700
Index first key cardinality: 18700
Index first 2 keys cardinality: -1
Index first 3 keys cardinality: -1
Index first 4 keys cardinality: -1
Index sequential pages: 0
Index page density: 0
Index avg sequential pages: 0
Index avg gap between sequences:0
Index avg random pages: 2026
Fetch avg sequential pages: -1
Fetch avg gap between sequences:-1
Fetch avg random pages: -1
Index RID count: 765027
Index deleted RID count: 0
Index empty leaf pages: 0
Base Table Schema: DOA
Base Table Name: ODS_SFC_SALE_BARCODE_T
Columns in index:
BILL(A)
Base Table For Index Not Already Shown:
---------------------------------------
Schema: DOA
Name: ODS_SFC_SALE_BARCODE_T
Time of creation: 2011-01-15-13.40.52.458517
Last statistics update: 2012-04-23-18.45.51.142593
Number of data partitions: 1
Number of columns: 21
Number of rows: 765027
Number of pages: 50996
Number of pages with rows: 50996
Table overflow record count: 0
Indexspace name: TS_INDEX_4K
Tablespace name: TS_4K
Tablespace overhead: 24.100000
Tablespace transfer rate: 0.900000
Prefetch page count: -1
Container extent page count: 32
Long tablespace name: TS_4K
(转)explain、db2exfmt 命令的使用:文本输出执行计划的更多相关文章
- MySQL 执行计划explain详解
MySQL 执行计划explain详解 2015-08-10 13:56:27 分类: MySQL explain命令是查看查询优化器如何决定执行查询的主要方法.这个功能有局限性,并不总会说出真相,但 ...
- 0912MySQL 执行计划explain详解
转自http://blog.itpub.net/29773961/viewspace-1767044/ 该博客内容是比较全的,虽然写的比较晦涩,多读几遍还是不错的 explain命令是查看查询优化器如 ...
- explain查看sql执行计划
http://www.cnblogs.com/wolf-sun/p/5291563.html 一该命令作用:该命令会向您展示查询是如何被执行的. 1.各个项的含义:https://blog.csdn. ...
- 【mysql】mysql 调优之 ——执行计划 explain
1.what is explain(explain 是个什么东东) explain(解释),在 Mysql 中 作为一个关键词,用来解释 Mysql 是如何执行语句,可以连接 select .dele ...
- 高性能mysql 附录D explain执行计划详解
EXPLAIN: extended关键字:在explain后使用extended关键字,可以显示filtered列和warning信息.在较旧的MySQL版本中,扩展信息是使用EXPLAIN EXTE ...
- MySQL执行计划【explain】详解
本文已经收录到github仓库,仓库用于分享Java相关知识总结,包括Java基础.MySQL.Springboot.mybatis.Redis.rabbitMQ等等,欢迎大家提pr和star! gi ...
- Mysql explain执行计划
EXPLAIN(小写explain)显示了mysql如何使用索引来处理select语句以及连接表.可以帮助选择更好的索引和写出更优化的查询语句. EXPLAIN + sql语句可以查看mysql的执行 ...
- 详解Linux中的cat文本输出命令用法
作系统 > LINUX > 详解Linux中的cat文本输出命令用法 Linux命令手册 发布时间:2016-01-14 14:14:35 作者:张映 我要评论 这篇 ...
- 使用MySQL中的EXPLAIN解释命令来检查SQL
我们看到许多客户的系统因为SQL及数据库设计的很差所以导致许多性能上的问题,这些问题不好解决,但是可以采用一套简单的策略来检查生产系统,发现并纠正一些共性问题. 很显然,您应该尽最大努力设计出最好的数 ...
随机推荐
- .Net实现Windows服务安装完成后自动启动的两种方法
考虑到部署方便,我们一般都会将C#写的Windows服务制作成安装包.在服务安装完成以后,第一次还需要手动启动服务,这样非常不方便. 方法一:在安装完成事件里面调用命令行的方式启动服务 此操作之前要先 ...
- nginx 下载 大文件被截断
如果出现大文件被截断,且ngix的日志大量出现以下类似报错: 则说明是nginx没有fastcgi_temp的读写权限.其中fastcgi_temp是自己的文件夹名称,每个人的不同且路径也会不同,这个 ...
- 启动Android App时,动态将Sqlite数据库文件导入到手机中类方法
package com.aqioo.db; import java.io.File; import java.io.FileOutputStream; import java.io.InputStre ...
- Mysql之数据表操作
数据表操作: 查看当前数据库中所有的表: show tables; 查看当前使用数据库: select database(); 使用数据表: use 表名; 创建数据表: create table 数 ...
- Ng第十二课:支持向量机(Support Vector Machines)(三)
11 SMO优化算法(Sequential minimal optimization) SMO算法由Microsoft Research的John C. Platt在1998年提出,并成为最快的二次规 ...
- Ng第一课:引言(Introduction)
Machine Learning(机器学习)是研究计算机怎样模拟或实现人类的学习行为,以获取新的知识或技能,重新组织已有的知识结构使之不断改善自身的性能. 它是人工智能的核心,是使计算机具有智能的根本 ...
- MapGis如何实现WebGIS分布式大数据存储的
作为解决方案厂商,MapGis是如何实现分布式大数据存储的呢? MapGIS在传统关系型空间数据库引擎MapGIS SDE的基础之上,针对地理大数据的特点,构建了MapGIS DataStore分布式 ...
- DeepFace和GAN
由于换脸技术的影响,现在造假视频的成本越来越低.AI换脸视频也越来越热门,甚至有一些已经达到了以假乱真的程度.虽然有明星反对表示无奈,可是.... 据报道,2018年,arXiv上发布了902篇GAN ...
- 20155326 《Java程序设计》第8周学习总结
20155326 <Java程序设计>第8周学习总结 教材学习内容总结 NIO (1)NIO使用频道来衔接数据节点,在处理数据时,NIO可以让你设定缓冲区容量,在缓冲区中对感兴趣的数据区块 ...
- Curl工具的使用
Curl命令可以通过命令行的方式,执行Http请求.在Elasticsearch中有使用的场景,因此这里研究下如何在windows下执行curl命令. 工具下载 在官网处下载工具包:http:// ...