Sravan Kumar

• Jan 9th, 2007

 

Source is changing Incrementally means the data in the source is keep on changing and your capturing that changes with time stamps,keyrange and with triggers,scds.Capture these changes to load the data incrementally.If We cannot capture this source data incrementally the data loading process will be very difficult.For example we have a source where the data is changing.On 09/01/07 we have loaded all the data in to the target.On 10/01/07 the source is updated with some new rows .It is very difficult to load all the rows again in the target,so what we have to do is capture the data whch is not loaded,and load only the changed rows into the target.

Srinivas Kumar

• Jan 22nd, 2007

 

I have a good example to explain this . Think about HR Data for a very big Company. The data keep on changing every minute. Now We had build a downstream system to capture a chunk of data for specific purpose say  "New Hirees". Every record in the source will have time stamp, when we load data today, we check the records which are updated/inserted today and will load them (This avoids to reprocess all the data).  We used incremental Refresh method to process such data. Infact most of the sources in OLTP are incremental changing/constantly changing.