Paged attributes can be used to have very large non-partioned table columns loaded into memory in smaller chunks. This comes to the expense of compression factor and scan efficency.
With HANA you see the idea is to keep all data you may want to work with in memory most of the time.
If you have lager amounts of data that you barely ever need, you may look into using LOB columns for that data.
Maybe it would be a good idea if you tell us what exactly it is you want to do - then its a lot easier for me to tell you if and how that's doable with HANA.