We describe the investigation of methods for cycle-accurate emulation of an Intel 8080 CPU using a low-power ARM11 CPU. Source-level optimizations and several threaded dispatch designs are explored, as well as an extreme optimization technique using direct register mapping and dynamic binary translation between the two Instruction Set Architectures. The cycle efficiency of the fastest design is nearly as fast as the original 8080 processor.

This paper presents an application that performs multi-objective geospatial optimizations for tactical mission planning in an urban environment. Utilizing a XML-driven C++ framework developed for hybrid platforms, the application distributes computational tasks to a collection of heterogeneous OpenCL devices and achieves efficient task parallel scaling. The framework abstracts the details of the mission scenario setup, compute device architecture details, and task scheduling to enable robust extensibility in software. Performance scalability and multi-objective parameter visualization for a prototypical mission scenario are documented in this study.

To move High-Performance Computing (HPC) closer to forward operating environments and missions, the Army Research Laboratory is developing approaches using hybrid, asymmetric core computing. By blending capabilities found in Graphics Processing Units (GPUs) and traditional von Neumann multicore Central Processing Units (CPUs), approaches are being developed and optimized to provide at or near real-time processing speeds for research project applications. Algorithms are designed to partition work to resources best designed to handle the processing load. The use of commodity resources allows the design to be flexible throughout the life cycle without the costly and time-consuming delays associated with Application-Specific Integrated Circuit (ASIC) development. This paradigm allows for rapid technology transfer to end users. In this paper, we describe a synchronous impulse reconstruction radar imaging algorithm that has been designed for hybrid CPU-GPU processing. We discuss various optimizations such as asynchronous task partitioning between the CPU and GPU as well as data movement reduction. We also discuss analysis and design of the algorithms within the context of two programming models: NVIDIA's CUDA and AMD's ATI Brook+. Finally, we report on the speedup achieved by this approach that allowed us to take a code once restricted to postprocessing and transform it into one that exceeds real-time performance requirements.