#: Brian Moseley changed the world a bit at a time by saying (astral date:
4/19/2006 7:07 PM) :#
On 4/19/06, Alexandru Popescu <[EMAIL PROTECTED]> wrote:
I am wondering if this problem may occur with Derby too?
yes, it does. derby's default blob size is 1M, iirc.
I would really like to better understand what may go wrong here. Please stop me at what point I am
going wrong:
1/ The blob is used to persist a NodeState. A NodeState's heavy part is the
ChildNodeEntries.
2/ ChildNodeEntries is mainly a double map of ChildNodeEntry
3/ a ChildNodeEntry is mainly NodeId.
4/ a NodeId is maily 2 long-s => 128bits
So, in order to meet the limitations of the DB schema, we should have a node that has as many
children as a serialized LinkedMap, HashMap of ChildNodeEntry will fill the length of the field.
Is this the correct computation?
./alex
--
.w( the_mindstorm )p.
