If I carry out a full extract of an InfoPackage which has 2LIS_13_VDITM there will be about 10 million records extracted (which currently not only takes time but there is an internal memory bug which is being fixed otherwise). There is already a delta load running for this extractor. I need to update 10 records, which fall into sets which can be represented by three document numbers (VBELN), which are 0040006506, 0040006502, 0040013065, I plan to do three separate loads for each of the load using VBELN in the selection criteris. Can anyone confirm that this is the best approach in this case. Thanks