Hi,
WE are establishing connectors from siebel database ( in
SQL 2000) to SQL data warehouse.
We are using microsoft DTS to build the connectors.
We have successfully established incremental update from
source to warehouse for inserted, updated records in
source.we have achieved this with the logic of comparing
the maximum updated date in source with that of DW.
However we do not know how to handle deleted records in
source for incremental update.These deleted records will
be hard deleted in source but they remain in DW bec the
logic looks for only maximum date for records that are
present in source database.
We cannot put triggers or any other mechanism in the
source database as it is not permissible.

Can someone suggest me how to handle incremental update
with maximum date comparison between source and target
tables which can accomodate deleted records also?
Would appreciate if u can provide some inputs and some
URLs publishing papers on incremental updates to Data
warehouse.
Regds
Krishna

It is important to realize you cannot incrementally
update deleted records. The cube MUST be full processed
if records have been deleted.

In order to handle this incrementally you will need to
take an accounting approach and negate the existing fact
as appropriate.

If you are deleting dimension members, try using a
deleted flag and coding to handle it that way.

Hope that helps,

Steve Hughes
Magenic Technologies