With the increased interest in the usage of GP-GPUs in conjunction with traditional CPUs in today’s high-end computing platforms, the task of developing consistent, well-written, and highly performing software to leverage all of the resources across these architectures has become a significant challenge. With the ever-increasing performance being produced by both styles of architecture, including increasing numbers of processing cores, creative caching and memory structures, and innovative new instructions, it has become imperative that a solution be developed between the developer community and the creators of these hardware devices to permit easy programming.