Overlapping computations and communication is a key to accelerating stencil applications on parallel computers, especially for GPU clusters. However, such programming is a time-consuming part of the stencil application development. To address this problem, we developed an automatic code generation tool to produce a parallel stencil application with latency hiding automatically from its dataflow model. With this tool, users visually construct the workflows of stencil applications in a dataflow programming model. Our dataflow compiler determines a data decomposition policy for each application, and generates source code that overlaps the stencil computations and communication (MPI and PCIe). We demonstrate two types of overlapping models, a CPU-GPU hybrid execution model and a GPU-only model. We use a CFD benchmark computing 19-point 3D stencils to evaluate our scheduling performance, which results in 1.45 TFLOPS in single precision on a cluster with 64 Tesla C1060 GPUs.

The Cell Broadband Engine (Cell/B.E.) is a novel multi-core microprocessor designed to provide high-performance processing capabilities for a wide range of applications. In this paper, we describe the world's first JPEG 2000 and Motion JPEG 2000 encoder on Cell/B.E. Novel parallelization techniques for a Motion JPEG 2000 encoder that unleash the performance of the Cell/B.E. are proposed. Our Motion JPEG 2000 encoder consists of multiple video frame encoding servers on a cluster system for high-level parallelization. Each video frame encoding server runs on a heterogeneous multi-core Cell/B.E. processor, and utilizes its 8 Synergistic Processor Elements (SPEs) for low-level parallelization of the time consuming parts of the JPEG 2000 encoding process, such as the wavelet transform, the bit modeling, and the arithmetic coding. The effectiveness of high-level parallelization by the cluster system is also described, not only for the parallel encoding, but also for scalable performance improvement for real-time encoding and future enhancements. We developed all of the code from scratch for effective multilevel parallelization. Our results show that the Cell/B.E. is extremely efficient for this workload compared with commercially available processors, and thus we conclude that the Cell/B.E. is quite suitable for encoding next generation large pixel formats, such as 4K/2K-Digital Cinema.

The cell broadband engine (CBE) is a novel multi-core microprocessor designed to provide compact and high-performance processing capabilities for a wide range of applications. Real-time image processing applications with parallelism for large amounts of data are good examples to demonstrate the unique capabilities of the CBE. In this paper, we describe the evaluation of the performance for image processing using wavelet transforms on CBE. Our results show that the CBE is extremely efficient in this processing compared with commercially available processors, and thus, we conclude that the CBE is quite suitable for next generation large pixel formats, such as 4K/2K-digital cinema

We introduce a new deign approach of panoramic movie system. The system, named PanoVi by the authors, captures a 360-degree surround movie from a viewpoint by using four wide-angle video cameras, then the video streams are encoded as a single stream of standard video format. At the viewing side, geometric correction and image mosaicking are performed in real-time by using commodity 3D graphics accelerator hardware equipped with personal computers. By mapping the mosaic on the inside wall in 3D space, the viewing system renders a partial view of arbitrary direction as a virtual environment. This approach is suitable for real-time transmission of a panoramic movie because this system uses a standard video format in transmission layers. We developed a prototype system based on Windows and DirectX8 environments. We confirmed full-rate interactive panoramic movie playback on commodity Pentium4 based PCs.

This paper introduces a new panoramic movie system that is applicable for real-time transmission of circle view movies over networks. The system consists of a capturing and a viewing system. The capturing system collects 4 video streams from different directions. After correcting geometric distortion caused by spherical aberration of the lens, these streams are combined into an intermediate video stream. The PanoVi panoramic video format is designed to be compatible with the NTSC and MPEG2 standards. It can be sent to a client system via network and video recording media. At the viewing system, the video stream is decomposed into the original 4 streams and mapped onto walls in 3-space. By indicating viewing direction, the application generates a movie output that appears as if a camera were observing the direction. We developed a prototype of the PanoVi panorama movie system using Pentium III-based PC and confirmed its performance.