Scenario
There are five databases, two of which are SQL servers and three are Oracle10g.
Target
Migrate the business data of the two SQL servers to the three Oracle databases according to the specific logic.
Kettle usage (basic)
Kettle installation and configuration
- You can download kettle installation on the official website.
- A database is required for kettle configuration. Kettle database stores metadata descriptions, tasks, and conversions required by kettle. Kettle's default login information is admin/admin.
- Follow the prompts to create an Oracle database connection.
Server connection.
Note the difference between Database Name and instance for SQL Server connection
Name.
Basic concepts of kettle
- A job is responsible for organizing [conversion] together to complete a job. Generally, we need to split a large task into several logically isolated jobs, when these jobs are completed, the task is completed.
- Conversion: defines the container for data operations. Data operations are a process from input to output of data. It can be understood as a container with a lower granularity than a job. We divide tasks into jobs, then, you need to split the job into one or more transformations, and each transformation completes only part of the work.
Kettle basic example
- Kettle's error handling requires error logging in many scenarios, for example, if the migration prompts data problems, primary/foreign key errors, or violation of constraints, the current scenario should be recorded in one place for special processing.
Example
Main Process
Error message Configuration
- When the data volume is large, you can add filter parameters for processing.
- Value ing
- Connection
Note that the raw data used for the connection must be sorted in ascending order.
References
- Baidu Library # bi-pentaho authoritative guide
- Http://www.pentaho.com/
- Http://blog.csdn.net/ybygjy/article/details/7478113
