分析IBM datastage job 构建过程

在IBM datastage job 中我们知道共有三种类型，在此简单阐述下

1.server job ，运行在server 服务器端，简单用于逻辑控制的job 可以采用这种方法实现

2.parallel job ，并行运行的job，对于大多数的job develop 采用这种方法来实现

3.sequence job，主要是作流程控制用的。

简单举例说明：

现在有一个parallel job which named edw_dss_dm_dang.这个job 可能不是很复杂，是用来进行数据加载或是备份的我们称之为(ETL过程)。

现在要把这个job 加入到一个单元 sequence中，也就说这个sequence 是专门为dang类job 操作的。we call this sequence as dang_sequence.before we added this dang job into dang sequence need to decide when the job should be run ,so 在这里我们添加一个参数 run_time to decide which dang job can run named dang_can_ run_time which setted to boolean type。when the parameter is Y then dang_job can run,but the parameter is set to default value N.SO when we run dang_sequence ，Have to manutal 。

but，as all we know ,there is a totally sequence ,which we called totally_sequence ,so the parameter in this job are good ,its mean every time we run the job ,each job in it

can run unless we do something .so there a parameter should be called dang_can_ run_time,and the parameter default valued not be set to N OR Y but default。 because we should sett it use program 。so we may use server job .

有一个sql 能够很好的说明这一点 :

礼拜一可以处理

select 'dang_can_ run_time' as pk, case when dayofweek(CURRENT DATE) in (2) then 'Y' 
else 'N' end as run_flag
from SYSIBM.SYSDUMMY1 d

这个数值被保存到hashed failed (named sns_run_flag_hash)中。in this failed there is only two columns named pk,run_flag。

but how can we get the values 

use a user variables activity just like this 

if dang_can_ run_time='DEFAULT' then UtilityHashLookup('sns_run_flag_hash', 'sns_run_flag_hash', 1) else sns_run_flag_hash.

then can user nested condition to decided run the job or not ，just so eary.