Sql Server性能优化方法备忘录

1. 30分钟内耗时最长20个Sql语句：
   SELECT TOP 20
   SUBSTRING(st.text, (qs.statement_start_offset/2) + 1,
   ((CASE statement_end_offset
   WHEN -1 THEN DATALENGTH(st.text)
   ELSE qs.statement_end_offset END
   - qs.statement_start_offset)/2) + 1) as SQL_Part --统计对应的部分语句

   , CAST( ((qs.total_elapsed_time / 1000000.0)/qs.execution_count) AS DECIMAL(28,2) ) AS [平均消耗秒数]
   , CAST(qs.last_elapsed_time / 1000000.0 AS DECIMAL(28, 2)) AS [最后完成消耗秒数]
   , qs.last_execution_time AS [最后执行时间]
   , CAST(qs.min_elapsed_time / 1000000.0 AS DECIMAL(28, 2)) AS [最小消耗秒数]
   , CAST(qs.max_elapsed_time / 1000000.0 AS DECIMAL(28, 2)) AS [最大消耗秒数]
   , CAST(qs.total_elapsed_time / 1000000.0 AS DECIMAL(28, 2)) AS [总消耗秒数]
   , (qs.execution_count) AS [总执行次数]
   , creation_time AS [编译计划的时间]
   , CAST(qs.last_worker_time / 1000000.0 AS DECIMAL(28, 2)) AS [最后完成占用CPU秒数]
   , qp.query_plan
   
   , st.text AS SQL_Full                                       --父级完整语句
   

   from sys.dm_exec_query_stats qs
   CROSS APPLY sys.dm_exec_sql_text(qs.sql_handle) AS st
   CROSS APPLY sys.dm_exec_query_plan(qs.plan_handle) qp

WHERE qs.last_execution_time>DATEADD(n,-30,GETDATE()) -- 30分钟内
ORDER BY qs.last_worker_time DESC

2. 查看 当前所有Session 及最后sql

   SELECT A.session_id, b.host_name,
   Db_Name(s.dbid) AS [database],
   DateDiff(MINUTE, A.connect_time, GetDate()) AS [Connected (mins)],
   num_reads, num_writes, login_name, q.text

   FROM sys.dm_exec_connections AS A
   INNER JOIN sys.dm_exec_sessions AS B ON A.session_id = B.session_id
   INNER JOIN sys.sysprocesses AS s ON s.spid = A.session_id
   cross APPLY sys.dm_exec_sql_text(sql_handle) q

注：当前活动对象
会话 sys.dm_exec_Session:
进程 sys.sysprocesses ： 跟session一样， 向后兼容用
连接 sys.dm_exec_connections： 底层tcp连接
请求 sys.dm_exec_requests

---

统计对象 sys.dm_exec_query_stats
sql语句 sys.dm_exec_sql_text(sql_handle)