Question

很抱歉这个模糊的问题。我只是不知道如何以一种简洁的方式进行解释。

我正在尝试运行将每个作业ID的job_modules分组的查询。问题在于表中有许多job_dis，而单个job_modules中有许多job_id。我只是不知道如何做一个小组。

这就是job_id的样子。表格中有多个job_id，但我想将模块分组。

|job_Module_____________|AVG_REC__|JOB_ID__
|DataSelectionProcess   |0.017    |32681
|DataSelectionProcess   |0.002    |32681
|DataSelectionProcess   |0        |32681
|DataSelectionProcess   |0.015    |32681
|DataSelectionProcess   |0.004    |32681
|DataSelectionProcess   |0.009    |32681
|DataSelectionProcess   |0.002    |32681
|DataSelectionProcess   |0.791    |32681
|DataSelectionProcess   |0.001    |32681
|DataSelectionProcess   |0        |32681
|DataSelectionProcess   |0.002    |32681
|DataSelectionProcess   |2.14     |32681
|OutputFileProcess      |1.926    |32681
|OutputFileProcess      |0.044    |32681
|OutputFileProcess      |0.001    |32681
|OutputFileProcess      |0.00305  |32681
|OutputFileProcess      |0.00102  |32681
|OutputFileProcess      |0        |32681
|OutputFileProcess      |0.00609  |32681
|RecallMatchCodeProcess |44.16148 |32681

以下是预期结果，应该将所有avg_rec时间加在一起，并添加所有模块：

|job_Module_____________|AVG_REC__|JOB_ID__
|DataSelectionProcess   |3.017    |32681
|OutputFileProcess      |2.006    |32681
|RecallMatchCodeProcess |44.16148 |32681

这是我要运行的查询。

Select  job_module, AVG_REC_TIME_MILLIS, job_id
        FROM job_step
        where job_id ='32681' or 
                job_id ='32687' or
                job_id ='32714' or
                job_id ='32716' or
                job_id ='32723' or
                job_id ='32743' or
                job_id ='32758' or
                job_id ='32767' or
                job_id ='32768' or
                job_id ='32782' or
                job_id ='32784' or
                job_id ='32800' or
                job_id ='32801' or
                job_id ='32849' or
                job_id ='32851' or
                job_id ='32884' or
                job_id ='32913' or
                job_id ='32928' or
                job_id ='32938' or
                job_id ='32947' or
                job_id ='33089' or
                job_id ='33095'
                group by job_module
                ORDER BY JOB_ID;

Answer 1

我认为您需要按job_id和job_module进行分组，如：

select job_id, job_module, sum(AVG_REC_TIME_MILLIS)
from job_step
where job_id in (
  '32681', '32687', '32714', '32716', '32723',
  '32743', '32758', '32767', '32768', '32782',
  '32784', '32800', '32801', '32849', '32851',
  '32884', '32913', '32928', '32938', '32947',
  '33089', '33095'
)
group by job_id, job_module

Answer 2

我认为AVG_REC 时间以秒和毫秒为单位。在这种情况下，您需要正确地汇总时间-通过将小数部分除以1000并用秒求和，将毫秒转换为秒。看起来像这样：

SELECT job_id, job_module, 
NUMTODSINTERVAL(SUM(
SUBSTR(avg_rec_time_millis, 1, instr(avg_rec_time_millis, '.')-1) +
(SUBSTR(avg_rec_time_millis, instr(avg_rec_time_millis, '.')+1)/1000)
)
, 'SECOND') sum_time
FROM job_step
WHERE job_id IN (
'32681', '32687', '32714', '32716', '32723','32743', '32758', '32767', '32768', '32782', '32784',
'32800', '32801', '32849', '32851','32884', '32913', '32928', '32938', '32947', '33089', '33095'
)
GROUP BY job_id, job_module;

您如何通过查询来创建一个子组？

2 个答案: