Insert: Generally this step is to add the new data (new changes) For Example, Sales Transaction for this week. [Before this, we have sales transaction data for past 3 months.]
Insert & Update: This is to filter new data from old data. so in the load statement, i put where last update > 11/4/2014 ( assuming that is the start of the date for the week). Then I concatenate this data to the old data. I also saw that someone suggested to use Where not Exists '11/3/2014' (assuming this is the ending date of old data- which is also the primary key..?).
Insert, Update & Delete: Use Inner Join with primary key. I do not understand this part. How does this delete redundant data, if the intention is to do so.
All of the above steps involves storing it in QVD, correct? I may have mixed up the contents in the above steps. if you can clarify the confusion, even blunder(possible!) I am making, that would be wonderful.
