OS: windows server 2008 R2 SE
We are implementing a new architecture for one our modules. There are 100 tables in the database and out of which 45 tables are using for reporting extensively. However I want to separate out to the tables used for reporting meaning I will create a
new database and put these 45 tables keep actual transactional database small and slim. So its like having a copy of the same tables but this copies will be used for reporting only from a different database. The retention of data in source tables is 1 day
where as the retention for the data in reporting(or destination) tables is an year and a half. so I was thinking of implementing replication based on a condition like based on a specific column value. But the issue here is when I run a purge job (maintenance)
on a nightly basis on the transactional tables how would i prevent the impact of not getting the data deleted from the reporting tables. The application that reads data from reporting tables needs the data to present all the time (cannot break replication
at any given point of time meaning data should get replicated continuously)
Appreciate your thoughts and inputs