The more times a piece of data is accessed in IPFS the more copies of it co-exist on the network. Therefore, if the data is useful in anyway then there will always be a copy of it residing on some node in the network.
Object pinning is used to describe when nodes ensure the survival of objects i.e. they may pay third parties to pin their object so that it’s always available.
White paper states:
3.5.3 Object Pinning Nodes who wish to ensure the survival of particular objects can do so by pinning the objects. This ensures the
objects are kept in the node’s local storage. Pinning can be done
recursively, to pin down all linked descendent objects as well. All
objects pointed to are then stored locally. This is particularly
useful to persist files, including references. This also makes IPFS a
Web where links are permanent, and Objects can ensure the survival of
others they point to.
Adding new data to IPFS doesn’t propagate to every node. Instead it broadcasts a message to the network stating that you have the data corresponding to a content address. If a node would like to access it then they can find it here. The data duplicated when another node requests it.
If you were sure that you would always have your nodes running to be able to 'serve' the data then I suppose could provide two copies yourself. Another option which I alluded to is to pay to have your data be persistent, using FileCoin.
Check out this video in which Juan Benet describes the possibilities of the technology. The Decentralized Web, IPFS and Filecoin
Hope this answers your question!