October 10, 2012
Held in conjunction with the PGAS12 Conference
Partitioned Global Address Space [PGAS] and shared memory programming models such as UPC, Co-Array Fortran, OpenSHMEM and OpenMP offer mechanisms to expose applications to globally shared memory and native thread/task parallelism. However, with the advent of heterogeneous computing architectures, there exists a significant gap in coupling PGAS-style languages to non-traditional computing devices such as GPU’s and reconfigurable processors. This workshop shall serve as a forum to present research developments and language-extensions for tightly coupled PGAS/shared memory languages and heterogeneous architectures.