您好,登錄后才能下訂單哦!
本篇內容介紹了“AWR TOP SQL實現方法是什么”的有關知識,在實際案例的操作過程中,不少人都會遇到這樣的困境,接下來就讓小編帶領大家學習一下如何處理這些情況吧!希望大家仔細閱讀,能夠學有所成!
1 按解析次數排序
select a.*, to_char(substr(b.sql_text,1,4000)) from (select dhs.sql_id, sum(parse_calls_delta) parse, sum(executions_delta) exec_nums, dhs.MODULE from dba_hist_sqlstat dhs where snap_id > 22438 and snap_id <= 22440 group by dhs.sql_id,MODULE) a, dba_hist_sqltext b where a.sql_id=b.sql_id order by a.parse desc;
2 按執行時間排序
select a.*, to_char(substr(b.sql_text,1,4000)) from (select dhs.sql_id, round(sum(elapsed_time_delta)/1000/1000,2) "elapsed_time(s)", sum(executions_delta) execs, round(sum(elapsed_time_delta)/1000/1000/sum(executions_delta),2) elapsed_time_per, dhs.MODULE from dba_hist_sqlstat dhs where snap_id > 22438 and snap_id <= 22440 group by dhs.sql_id,MODULE) a, dba_hist_sqltext b where a.sql_id=b.sql_id order by a."elapsed_time(s)" desc;
3 按CPU時間排序
select a.*, to_char(substr(b.sql_text,1,4000)) from (select dhs.sql_id, round(sum(cpu_time_delta)/1000/1000,2) "cpu_time", sum(executions_delta) execs, round(sum(cpu_time_delta)/1000/1000/sum(executions_delta),2) cpu_time_per, round(sum(elapsed_time_delta)/1000/1000,2) "elapsed_time(s)", dhs.MODULE from dba_hist_sqlstat dhs where snap_id > 22438 and snap_id <= 22440 group by dhs.sql_id,MODULE) a, dba_hist_sqltext b where a.sql_id=b.sql_id order by a."cpu_time" desc;
4 按User I/O wait排序
select a.*, to_char(substr(b.sql_text,1,4000)) from (select dhs.sql_id, round(sum(iowait_delta)/1000/1000,2) "iowait_time(s)", sum(executions_delta) execs, round(sum(iowait_delta)/1000/1000/sum(executions_delta),2) iowait_time_per, round(sum(elapsed_time_delta)/1000/1000,2) "elapsed_time(s)", dhs.MODULE from dba_hist_sqlstat dhs where snap_id > 22438 and snap_id <= 22440 group by dhs.sql_id,MODULE) a, dba_hist_sqltext b where a.sql_id=b.sql_id order by a."iowait_time(s)" desc;
5 按邏輯讀(gets)排序
select a.*, to_char(substr(b.sql_text,1,4000)) from (select dhs.sql_id, round(sum(buffer_gets_delta),2) "buffer_ges", sum(executions_delta) execs, round(sum(buffer_gets_delta)/sum(executions_delta),2) iowait_time_per, round(sum(elapsed_time_delta)/1000/1000,2) "elapsed_time(s)", dhs.MODULE from dba_hist_sqlstat dhs where snap_id > 22438 and snap_id <= 22440 group by dhs.sql_id,MODULE) a, dba_hist_sqltext b where a.sql_id=b.sql_id order by a."buffer_ges" desc;
7 按物理讀(physical read)排序
select a.*, to_char(substr(b.sql_text,1,4000)) from (select dhs.sql_id, round(sum(DISK_READS_DELTA),2) "physical_read", sum(executions_delta) execs, round(sum(DISK_READS_DELTA)/sum(executions_delta),2) iowait_time_per, round(sum(elapsed_time_delta)/1000/1000,2) "elapsed_time(s)", dhs.MODULE from dba_hist_sqlstat dhs where snap_id > 22438 and snap_id <= 22440 group by dhs.sql_id,MODULE) a, dba_hist_sqltext b where a.sql_id=b.sql_id order by a."physical_read" desc;
8 按執行次數排序
select a.*, to_char(substr(b.sql_text,1,4000)) from (select dhs.sql_id, round(sum(executions_delta),2) "exec_num", sum(ROWS_PROCESSED_DELTA) row_process, round(sum(ROWS_PROCESSED_DELTA)/sum(executions_delta),2) rows_per_exec, round(sum(elapsed_time_delta)/1000/1000,2) "elapsed_time(s)", dhs.MODULE from dba_hist_sqlstat dhs where snap_id > 22438 and snap_id <= 22440 group by dhs.sql_id,MODULE) a, dba_hist_sqltext b where a.sql_id=b.sql_id order by a."exec_num" desc;
“AWR TOP SQL實現方法是什么”的內容就介紹到這里了,感謝大家的閱讀。如果想了解更多行業相關的知識可以關注億速云網站,小編將為大家輸出更多高質量的實用文章!
免責聲明:本站發布的內容(圖片、視頻和文字)以原創、轉載和分享為主,文章觀點不代表本網站立場,如果涉及侵權請聯系站長郵箱:is@yisu.com進行舉報,并提供相關證據,一經查實,將立刻刪除涉嫌侵權內容。