Our Export API gives customers access to potentially large sets of data that would otherwise be difficult or slow to retrieve with our standard API. To support some of the tasks our standard API completes, the data has to be buffered in memory before it's returned. As a result, very large data sets can blow out the memory on our servers and destroy the API call that's making it.
What about paging?
We don't support paging. Anyone who has used MySql (and many other RDBMSes) knows that paging through a huge result set can be slow, especially when you start joining lots of tables. Paging is also a hassle when you just want "everything," which is the point of this API.