Well nested data parallelism generally applies to hierarchical structures like trees, whereas flat data parallelism applies to things like arrays and whatnot. Y

Well nested data parallelism generally applies to hierarchical structures like trees, whereas flat data parallelism applies to things like arrays and whatnot. You might want to watch this (similar in some ways) talk by SPJ on Data Parallel Haskell http://youtu.be/NWSZ4c9yqW8 http://youtu.be/NWSZ4c9yqW8