I have an application that is divided by sector (purchases, sales, inventory). I divide into different files by using the tab "Reduce" and grant access using the tab Distribute > Loop field in document.
The resulting files are small, but how they stay in memory? This is a better approach than reduction in Section Access?
It will depend widely on how the document has been built and how many resources does it use when in memory. Generally speaking, many small documents with little timeout values will perform better than one huge document accessed concurrently by a lot of users with higher timeouts. Hence, your approach seems all right to me, but don't take my word for it, and test it using several users and different sizes of documents.
Are your users going to make always the same queries, so they are going to be cached quickly and so returned faster (one huge doc) or are they rather going to make very different selections on many different fields and sets of information (many small docs)? How long does it take for them the use of the QlikView app?
.png "Miguel_Angel_Baeyens"){kind=avatar}
