Re: MT Design Question

This is a threading-related design question assuming that
C++0x is the implementation language.

Suppose I have a graph that I want to search for a node that
has some characteristic (e.g., holds a value within some
range). Suppose further that of the many ways to search the
graph (e.g., depth-first, breadth-first, random walk, etc.),
I can't predict which will be best, so I want to run them
all concurrently, stopping when either one of them finds a
suitable node or they all fail. That is, given something
like this

  Node* dfsSearch(Graph g, Predicate p); // do dfs search of g for
                                            // a node satisfying p
  Node* bfsSearch(Graph g, Predicate p); // do bfs search of g
  Node* rwSearch(Graph g, Predicate p); // do random walk search
of g
I want to do this:

  concurrently invoke dfsSearch, bfsSearch, and rwSearch;
  wait until one returns success or they all return lack of success;
  if one returned success, tell the others to stop searching;

One of the problems in mapping this to C++0x is that C++0x
has no notion of thread cancellation/interruption, but that
can be built manually, as Anthony Williams shows in section
9.2 of his "C++ Concurrency in Action"
(http://www.manning.com/williams/), which I'm pleased to be
able to plug here.
My question, however, has to do with implementing this part:

  wait until one returns success or they all return lack of success;