The nodes in the data structure are named using Path s, a doubly-linked list of integers that represent the branches. Each processor stores a hash table linking the paths to the global pointers that describe to which processor a node belongs.
