Multi-Tiled Architecture
MIT Raw and the like.

Mapping and Performance of Kernel Benchmarks for Tiled Architectures
James Lebak, Matt Alexander, Hector Chan, Ryan Haney, Preston Jackson, and Edmund Wong. In GOMAC 2005.

Parallel Computing
Taking advantage of multi-{core,tile,threaded} architectures.

pMapper: Automatic Mapping of Parallel Matlab Programs
Nadya Travinin, Henry Hoffmann, Robert Bond, Hector Chan, Jeremy Kepner, and Edmund Wong. In DoD HPCMP UGC 2005.