Does the framework guarantee no missed changes if a sync (between data stored in databases) occurs whilst updates to the source database are being made in another transaction that has not yet been committed?
I think that the only way to guarantee this is if the Knowledge sent from the Destination to the Source contains (summary) details about each and every Item that the Destination knows about. Is that correct? If so this could mean that the Knowledge sent is large.
In our case we need to sync up to 100,000 items. Initially the number of items will be relatively small (a few hundred to a few thousand). As the list of items grows there will be a lot of changes to sync but over time the items become more static so that much fewer changes need to be synced. After a while the Item Data sent from the Source to the Destination will be small though Knowledge about the 100,000 items still needs to be sent, right? There may be a few thousand users of our application so I am concerned about the perfomance even with relatively static data.
Cheers,
Jarrod
