博客
关于我
强烈建议你试试无所不能的chatGPT,快点击我
Notes: CRM Analytics–BI from a CRM perspective (2)
阅读量:5041 次
发布时间:2019-06-12

本文共 2852 字,大约阅读时间需要 9 分钟。

  1. Data Transfer Process
    1. As of NetWeaver 7.0 and later, Data Transfer Process replaces the function of data mart interface and the InfoPackage, so InfoPackage only loads the data into BI's entry level – PSA data.
    2. The transfer process can be implemented in parallel mode as it is set.
    3. Data load request is an instance of Data Transfer Process generated when in real time process, the request is executed according the settings of the DTP.
    4. Either full transfer or delta transfer can be implemented in DTP, In full mode, the entire dataset of the source is transferred to the target; in delta mode, only the data that was posted to the source since the last data transfer is transferred.
    5. DTP also has the function for handling with errors that happen during the data loading.
      1. At runtime, the incorrect data records are sorted and written to an error stack (request-based database table).
      2. A special error DTP will update the error stack info to the target further.
      3. Restart of the failed transfer and the search of the error processes are all very easy.
      4. Process chain is recommended.  
  2. Error DTP
    1. An error DTP is created for an activated data transfer process on the Update tab page, and runs directly in the background or be included in a process chain so that we can schedule it regularly in the context of our process chain.
    2. An error DTP uses a full update mode to transfer the error data, and reactions to the errors can be:
      1. No update, no reporting
      2. Updating valid records, no reporting (request red)
      3. Updating valid records, reporting possible
      4. Specify the maximum number of incorrect data records allowed before the transfer process is terminated.
      5. Select how the system should react when the number of data records received is not the same as the number of data records updated.
    3. Semantic Groups, the key fields for the error stack, can be defined in the "Execution" tab. The key fields need to be specified as much as possible.
    4. An error DTP can be defined and included in a process chain and started manually if error occurs.  
  3. Error Stack.
    1. An Error Stack is a request-based table into which the incorrect data records from a data transfer process are written.

            

    1.  The error data in the error stack can be post to the target manually, if this step is executed successfully, the error data will be deleted in the error stack. Also, if a request is deleted in a DataStore object, the associated data records are also deleted in the error stack.
  4.  Temporary Storage

    Temporary storage is a table that contains the data records processed with a request in a step together with a status display.

  5.  Extraction Monitor

    The request status and request messages from the data transfer processes (DTPs) and the InfoPackage load processes are integrated in the Extraction Monitor. (Administration -> Monitor)

转载于:https://www.cnblogs.com/rabbitzhang/archive/2009/02/10/1387372.html

你可能感兴趣的文章
一个小的日常实践——高速Fibonacci数算法
查看>>
创建与删除索引
查看>>
java的基本数据类型
查看>>
机器学些技法(9)--Decision Tree
查看>>
静态页面复习--用semantic UI写一个10min首页
查看>>
在Windows下安装64位压缩包版mysql 5.7.11版本的方法
查看>>
drf权限组件
查看>>
输入月份和日期,得出是今年第几天
查看>>
利用mysqldump备份mysql
查看>>
Qt中子窗口全屏显示与退出全屏
查看>>
使用brew安装软件
查看>>
[BZOJ1083] [SCOI2005] 繁忙的都市 (kruskal)
查看>>
吴裕雄 python 机器学习——数据预处理嵌入式特征选择
查看>>
Centos6.4安装JDK
查看>>
201521123069 《Java程序设计》 第4周学习总结
查看>>
线性表的顺序存储——线性表的本质和操作
查看>>
【linux】重置fedora root密码
查看>>
用swing做一个简单的正则验证工具
查看>>
百度坐标(BD-09)、国测局坐标(火星坐标,GCJ-02)和WGS-84坐标互转
查看>>
pig自定义UDF
查看>>