implementing large trees

Paul Rubin phr-n2002b at NOSPAMnightsong.com
Mon Jul 15 22:18:26 EDT 2002


How much info do you need to access at each node?

If not too much info per node, your best bet is probably figure out a
compact in-memory representation for the tree, preferably organized
for reasonable paging locality during traversal.  Then use the array
and/or mmap modules to crunch on the in-memory structure, decoding
individual nodes and edges as necessary.  With luck, this will fit
inside a few hundred megabytes of memory, which is not that large an
amount for a workstation these days.




More information about the Python-list mailing list