性能比较工具runstats
runstats能对做同一件事的两个不同方法进行比较,得出谁好一点。我们只需要提供两个不同方法,余下的事情都由runstats负责。runstats只是测量3个要素:
1. 耗用的时间。
2. 系统统计结果:会并排地显示每个方法做某件事(如执行一个解析调用)的次数,并展示二者之差。
3. latching,这是报告关键的输出。
安装过程:
create global temporary table run_stats ( runid varchar2(15), name varchar2(80), value int ) on commit preserve rows; create or replace view stats as select 'STAT...' || a.name name, b.value from v$statname a, v$mystat b where a.statistic# = b.statistic# union all select 'LATCH.' || name, gets from v$latch union all select 'STAT...Elapsed Time', hsecs from v$timer; create or replace package runstats_pkg as procedure rs_start; procedure rs_middle; procedure rs_stop( p_difference_threshold in number default 0 ); end; / create or replace package body runstats_pkg as g_start number; g_run1 number; g_run2 number; procedure rs_start is begin delete from run_stats; insert into run_stats select 'before', stats.* from stats; g_start := dbms_utility.get_time; end; procedure rs_middle is begin g_run1 := (dbms_utility.get_time-g_start); insert into run_stats select 'after 1', stats.* from stats; g_start := dbms_utility.get_time; end; procedure rs_stop(p_difference_threshold in number default 0) is begin g_run2 := (dbms_utility.get_time-g_start); dbms_output.put_line ( 'Run1 ran in ' || g_run1 || ' hsecs' ); dbms_output.put_line ( 'Run2 ran in ' || g_run2 || ' hsecs' ); if ( g_run2 <> 0 ) then dbms_output.put_line ( 'run 1 ran in ' || round(g_run1/g_run2*100,2) ||'% of the time' ); end if; dbms_output.put_line( chr(9) ); insert into run_stats select 'after 2', stats.* from stats; dbms_output.put_line ( rpad( 'Name', 30 ) || lpad( 'Run1', 12 ) ||lpad( 'Run2', 12 ) || lpad( 'Diff', 12 ) ); for x in ( select rpad( a.name, 30 ) || to_char( b.value-a.value, '999,999,999' ) ||to_char( c.value-b.value, '999,999,999' ) || to_char( ( (c.value-b.value)-(b.value-a.value)), '999,999,999' ) data from run_stats a, run_stats b, run_stats c where a.name = b.name and b.name = c.name and a.runid = 'before' and b.runid = 'after 1' and c.runid = 'after 2' and abs( (c.value-b.value) - (b.value-a.value) )> p_difference_threshold order by abs( (c.value-b.value)-(b.value-a.value)) ) loop dbms_output.put_line( x.data ); end loop; dbms_output.put_line( chr(9) ); dbms_output.put_line ( 'Run1 latches total versus runs -- difference and pct' ); dbms_output.put_line ( lpad( 'Run1', 12 ) || lpad( 'Run2', 12 ) || lpad( 'Diff', 12 ) || lpad( 'Pct', 10 ) ); for x in ( select to_char( run1, '999,999,999' ) || to_char( run2, '999,999,999' ) || to_char( diff, '999,999,999' ) || to_char( round( run1/decode( run2, 0, to_number(0), run2) *100,2 ), '99,999.99' ) || '%' data from ( select sum(b.value-a.value) run1, sum(c.value-b.value) run2, sum( (c.value-b.value)-(b.value-a.value)) diff from run_stats a, run_stats b, run_stats c where a.name = b.name and b.name = c.name and a.runid = 'before' and b.runid = 'after 1' and c.runid = 'after 2' and a.name like 'LATCH%' ) ) loop dbms_output.put_line( x.data ); end loop; end; end; /
试验设计,测试对比加不加绑定变量的情况:
--测试过程的sql drop table t purge; create table t ( x int ); --无绑定变量 create or replace procedure p1 as l_cnt number; begin for i in 1 .. 10000 loop execute immediate 'select count(*) from t where x = ' || i into l_cnt; end loop; end; / --绑定变量 create or replace procedure p2 as l_cnt number; begin for i in 1 .. 10000 loop select count(*) into l_cnt from t where x = i; end loop; end; / exec runStats_pkg.rs_start; exec p1; exec runStats_pkg.rs_middle; set serveroutput on; exec p2; exec runStats_pkg.rs_stop(1000);
测试结果:
Run1 ran in 311 hsecs
Run2 ran in 30 hsecs
run 1 ran in 1036.67% of the time
Name Run1 Run2 Diff
STAT...bytes sent via SQL*Net 276 1,524 1,248
LATCH.session allocation 1,695 74 -1,621
STAT...parse count (hard) 10,011 5 -10,006
STAT...enqueue releases 10,016 8 -10,008
STAT...enqueue requests 10,017 8 -10,009
STAT...parse count (total) 10,041 14 -10,027
STAT...calls to get snapshot s 50,073 40,020 -10,053
STAT...consistent gets 40,194 30,060 -10,134
STAT...consistent gets from ca 40,194 30,060 -10,134
STAT...session logical reads 40,232 30,087 -10,145
STAT...recursive calls 20,920 10,076 -10,844
LATCH.enqueue hash chains 20,117 16 -20,101
LATCH.enqueues 20,124 18 -20,106
LATCH.cache buffers chains 80,492 60,209 -20,283
STAT...physical read total byt 0 24,576 24,576
STAT...physical read bytes 0 24,576 24,576
LATCH.kks stats 25,415 10 -25,405
LATCH.library cache pin 70,385 20,185 -50,200
STAT...session uga memory 65,560 7,488 -58,072
LATCH.library cache lock 60,292 179 -60,113
STAT...session pga memory 65,536 0 -65,536
LATCH.row cache objects 150,201 356 -149,845
LATCH.shared pool 206,050 10,230 -195,820
STAT...session uga memory max 261,964 65,560 -196,404
LATCH.library cache 216,965 20,474 -196,491
STAT...session pga memory max 262,144 65,536 -196,608
Run1 latches total versus runs -- difference and pct
Run1 Run2 Diff Pct
854,415 111,834 -742,581 764.00%
相关文章
- Linux 性能调优CPU篇:工具命令篇
- Python 性能测试工具Locust
- Python性能分析工具Profile
- JMeter性能测试介绍学习一
- 【性能优化】CBO,RBO在ORACLE中的应用
- Chrome网页性能分析工具
- Seata 1.6.0 正式发布,大幅度提升存储性能
- SQLServer性能数据解析
- Traceview 性能分析工具
- Atitit 提升扩展性的艺术 艾提拉著 目录 1. . 功能扩展 vs 性能扩展1 1.1. udf机制 sp 等插件机制1 1.2. 1.5. 脚本。脚本是扩展复杂功能的利器21 1.3
- paip.提升性能---jvm java 工具使用.
- 单站验证概述(工具介绍、指标性能解释)
- 移除 GIL,可显著提升 Python 多线程性能么?
- 【Java】java 性能监控及工具
- 性能优化方法论举例(***)
- 干货:如何写性能测试计划
- 测试开发 | JMeter 典型电商场景(下单/支付)的性能压测
- web页面性能检测工具Lighthouse
- Memcache 内存分配策略和性能(使用)状态检查
- NILM笔记:性能度量(Performance Metrics)
- linux 内存性能评估工具 free/smem
- Linux网络安全运维:网络性能评估工具iperf 测试带宽
- NAS 性能问题处理
- Oracle的学习心得和知识总结(十八)|Oracle数据库性能压测工具swingbench的安装和使用及AWR ASH ADDM报告生成
- 利用shell脚本写一个系统性能分析工具
- 自动驾驶辅助系统性能评估工具MXeval4.1版本更新快讯