http://qs1969.pair.com?node_id=232461


in reply to Re: A memory efficient hash, trading off speed - does it already exist?
in thread A memory efficient hash, trading off speed - does it already exist?

Can you explain to me why the structures that store the data take over 10 times the size of the actual data itself? I believe you, I just don't understand why.
What does storing the data in this format provide? Speedy STOREs and FETCHes? Is that it?

I have a friend busily telling me that if this was C he could simply construct a linked-list struct and wander across the structs, which would undoubtedly take up less memory. While I'm sure thats not as magically efficient as he thinks, and I know the lookups would be phenomenally slow, its hard to convince him :)

Thanks,
JP
-- Alexander Widdlemouse undid his bellybutton and his bum dropped off --