Say I have Parent class and Parent has getChildren, which is mapped using a one-to-many set mapping.

If my Children are all very large blob objects and I query to getParent, does this pull down all the children as well? As per my understanding, it would, but I am a major hibernate newbie.

And if so, what is the best way to accomplish such a mapping without having to pull down all the large blob children, but rather only get them on demand? Is it better to store just the ids and somehow store the blob children in seperate tables?

thank you.