On Oct 30, 2009, at 9:14 AM, Biddiscombe, John A. wrote:
If I have a representation of a tree and I have sub-portions of it spread across a number of processes (but the whole tree represents a well-formed and simple tree structure). Is there a reasonably straightforward way of doing this using BGL or PBGL. I had hoped that it might be as simple as defining some templated tree structure and using the add operator after gathering pieces onto one node.
(The tree structure could be represented as XML where the root would exist on all processes, different sub-nodes and leaves and would occur on different processes - a simple xml merge would probably suffice for my needs right now).
If examples using boost libraries ((p)BGL/MPI) or other similar combination exist and anyone can point me to resources I'd be most grateful. (I'm not an expert with BGL so will need some do some learning...)
Thanks
JB
I'm unclear exactly what your merge operation/algorithm is (possibly because I never use XML which probably provides adequate context for other people). The p(BGL) graph classes should work fine for representing your tree. If you describe your algorithmic needs in more detail I can suggest whether PBGL or BGL with a bit of MPI for the data movement is more appropriate. I presume the total structure is small enough to represent on a single node since you seem to be doing an all_gather? Thanks, Nick