在流水线结构上的基于Shear-Wrap的并行体数据绘制算法

在以前的基于目标空间划分的并行体数据绘制算法中，局部绘制和图象融合是两个串行的过程，在节点机的局部绘制阶段几乎没有数据通讯，但在数据融合阶段数据通讯量非常大，出现总线争用甚至通讯阻塞，而且在这个阶段有非常大的同步开销。本文利用流水线结构，让局部体数据绘制和图象融合并行执行，很好地解决了上述缺点。并在一个基于微机的流水线结构上实现了一个新的基于目标空间划分的并行体数据绘制算法。     

并行文件系统中高效遥感数据分布技术之研究

本文简要回顾了并行文件系统领域方面的研究,介绍了一种高效的基于大数据量遥感信息的并行文件系统PIPFS(ParallelRemoteSensingImageProcessingFileSystem)中的数据分布机制和元数据格式,说明了该并行文件系统设计中数据分布的主要特点,并对系统特有的数据分布方式和元数据格式进行了详细说明,进行了相关数据测试,用以说明该并行文件系统利用数据分布策略来提高效率的独特特点。     

OpenMP数据分布子句自动生成算法

将OpenMP程序扩展到异构多核结构时,非本地存储访问会导致访存开销增加,影响程序性能。针对该问题,引入带数组划分信息的数据分布子句,对数据在异构多核存储系统的布局进行管理,提出一种基于并行循环识别和数组引用模式分析的算法,实现该类子句的自动生成。实验结果表明,自动生成的OpenMP程序包含数据分布子句,具有较好的数据局部性,可降低访存开销,在异构多核系统上获得明显的性能提升。     

并行文件系统中高效遥感数据分布技术之研究

本文简要回顾了并行文件系统领域方面的研究,介绍了一种高效的基于大数据量遥感信息的并行文件系统PIPFS(Parallel Remote Sensing Image Processing File System)中的数据分布机制和元数据格式,说明了该并行文件系统设计中数据分布的主要特点,并对系统特有的数据分布方式和元数据格式进行了详细说明,进行了相关数据测试,用以说明该并行文件系统利用数据分布策略来提高效率的独特特点.     

基于图像的虚拟场景实时漫游

提出一种基于图像的并行绘制算法 ,通过分割图像显示窗口并行绘制 ,实现了三维场景的高分辨率图像的实时绘制 该算法按各节点计算性能分配任务 ,实现了负载平衡 利用三维场景漫游时相邻帧之间的相关性 ,采用RLE(RunLengthEncoded)压缩方法传输图像 ,降低了传输开销 实验结果表明 ,在 4台PC组成的机群上 ,实现了5 12× 5 12大小的图像的实时绘制     

基于PVM的OpenGL的并行实现

本文在以微机为节点的并行虚拟机(PVM)上实现了三维图形工业标准OpenGL,具有实用价值.文中提出了新的图像平面分割算法和图像区域的分块绘制与回传方法,使得各处理机之间的数据交换量较小且分布在整个绘制过程中.该方法特别适合于通信开销大的网络分布式计算环境.利用帧间一致性和线性预测,使得任务的分配与绘制并行完成,避免了等待.本文的研究结果可直接应用于建立基于微机网络的交互式实时三维图形系统.     

大规模数据密集型系统中的去重查询优化

针对shared-nothing结构下大规模数据密集型系统去重查询的挑战,提出了一种有效的数据分布策略和并行处理方法分别对相关属性和无关属性去重进行优化:即自适应的散列和直方图相结合的数据分布策略,以及异步式并行查询中间件.前者在数据写入时保证数据均衡,并在数据量发生倾斜时自动调整数据的分布;后者充分发掘了去重查询处理中的粗粒度流水级并行,并消除了多节点同步等待的开销,尽早地返回结果.在生产系统DBroker上的测试表明,数据分布策略极大地改善相关属性的去重查询性能,而异步式并行查询引擎能够充分发掘并行性,对不相关属性的去重查询具有显著的性能提升.     

并行重构程序中过程调用的优化技术

过程间分析技术是提高编译器目标码效率的重要技术。在分布式存储的并行系统中,数据分布的好坏对应用程序的性能有重要影响。但是仅做过程内的数据对齐和分布推导难以获得良好的全局数据分布模型,而且会使过程调用的开销难以控制,因而如何处理过程调用,以及数据分布信息在过程调用点的收集与传播对生成高效率的目标码是极其重要的。文中应用过程调用实实例化技术,减小过程调用的开销,并能很好地满足数据局部性要求。     

MPI自动并行化编译系统中消息传递代码生成算法

传统MPI自动并行化编译系统从数据重分布的角度,生成面向分布式存储系统的消息传递程序,但是大量数据重分布通信的额外开销导致其加速比低。为了解决此问题,在基于Open64的MPI自动并行化编译系统后端,提出了一种消息传递代码生成算法。该算法以统一数据分布为中心,根据给定的并行化循环集和通信数组集,通过修改WHIRL表示的串行代码语法结构树,生成更精确的消息传递代码。实验结果表明,该算法能够较大程度地降低消息传递程序的通信开销,并且明显提升其加速比。     

分布式并行绘制系统中带宽降低方法的研究

把静态与视点无关网格简化技术和动态与视点相关的网格简化技术结合起来，根据应用领域的要求，采用考虑了离散曲率边折叠算法构造的多分辨率模型，把得到的简化模型提交给与视点相关的网格简化技术处理，在基于联网PC的分布式并行绘制系统中实现模型的渐进传输并进行实时绘制；同时采用几何索引等方式来降低立即模式分布式系统的网络传输数据量，进一步缓解网络带宽压力，增强数据的实时动态交互性。     

A low-overhead networking mechanism for virtualized high-performance computing systems

The use of virtualized parallel and distributed computing systems is rapidly becoming the mainstream due to the significant benefit of high energy-efficiency and low management cost. Processing network operations in a virtual machine, however, incurs a lot of overhead from the arbitration of network devices between virtual machines, inherently by the nature of the virtualized architecture. Since data transfer between server nodes frequently occurs in parallel and distributed computing systems, the high overhead of networking may induce significant performance loss in the overall system. This paper introduces the design and implementation of a novel networking mechanism with low overhead for virtualized server nodes. By sacrificing isolation between virtual machines, which is insignificant in distributed or parallel computing systems, our approach significantly reduces the processing overhead in networking operations by up to 29% of processor load, along with up to 36% of processor cache miss. Furthermore, it improves network bandwidth by up to 8%, especially when transmitting large packets. As a result, our prototype enhances the performance of real-world workloads by up to 12% in our evaluation.     

Equalizer: A Scalable Parallel Rendering Framework

Continuing improvements in CPU and GPU performances as well as increasing multi-core processor and cluster-based parallelism demand for flexible and scalable parallel rendering solutions that can exploit multipipe hardware accelerated graphics. In fact, to achieve interactive visualization, scalable rendering systems are essential to cope with the rapid growth of data sets. However, parallel rendering systems are non-trivial to develop and often only application specific implementations have been proposed. The task of developing a scalable parallel rendering framework is even more difficult if it should be generic to support various types of data and visualization applications, and at the same time work efficiently on a cluster with distributed graphics cards. In this paper we introduce a novel system called Equalizer, a toolkit for scalable parallel rendering based on OpenGL which provides an application programming interface (API) to develop scalable graphics applications for a wide range of systems ranging from large distributed visualization clusters and multi-processor multipipe graphics systems to single-processor single-pipe desktop machines. We describe the system architecture, the basic API, discuss its advantadges over previous approaches, present example configurations and usage scenarios as well as scalability results.     

复式并行流水线在基于PC集群机的并行绘制中的应用

提出基于动态绘制组的混合式体系结构,除了动态绘制组间的并行处理流水线外,在动态绘制组内部设计了缓帧并行流水线改进工作流程,形成了复式的并行绘制流水线,大大地提高了基于PC集群机的并行图形绘制系统的整体性能.采用此复式流水线的原型系统在实际测试中表现突出,性能比单层并行绘制流水线绘制系统有较大提高.     

Dynamic load balancing for parallel polygon rendering

Using parallel processing for visualization speeds up computer graphics rendering of complex data sets. A parallel algorithm designed for polygon scan conversion and rendering is presented which supports fast rendering of highly complex data sets using advanced lighting models. Dedicated graphics rendering engines do not necessarily suit such data sets, although they can support real-time update of moderately complex scenes using simple lighting. Advantages to using a software-based approach include the feasibility of adding special rendering features to the program and the capability of integrating a parallel scientific application with a parallel graphics renderer. A new work decomposition strategy presented, called task adaptive, is based on dynamically partitioning the amount of computational work left at a given time. The algorithm uses a heuristic for dynamic task decomposition in which image space tasks are partitioned without requiring interruption of the partitioned processor. A sophisticated memory referencing strategy lets local memory access graphics data during rendering. This permits implementation of the algorithm on a distributed memory multiprocessor. An in-depth analysis of the overhead costs accompanying parallel processing shows where performance is adequate or could be improved     

基于面向服务对象体系结构的交通噪声地图分布式计算方法

针对目前城市交通噪声地图绘制工具和方法不能适应大规模项目分布式计算需要的现状,提出一种基于松耦合服务的噪声地图分布式计算方法,阐述了面向服务对象体系结构(SOOA)运作机制,给出了噪声地图计算服务的建立方法并介绍了其部署方式和管理模式。最后通过某示范区噪声地图求解实例验证了该算法在有效提高计算效率的同时能够给系统提供充分的灵活性。实验结果表明,并行子任务的不均衡会影响并行效率,一般状况下并行效率能够达到85%以上。     

Large scale distributed visualization on computational Grids: A review

Advances in science and engineering have put high demands on tools for high performance large-scale data exploration and analysis. Visualization is a powerful technology for analyzing data and presenting results. Todays science and engineering have benefited from state-of-the-art of Grid technologies and modern visualization systems. To visualize the large amount of data, rendering technologies are widely used to parallelize visualization tasks over distributed resources on computational Grids. It raises the necessity to balance the computational load and to minimize the network bandwidth requirements. This article explains in Grid environments how new approaches of visualization architecture and load-balancing algorithms address these challenges in a principled fashion. The Grid infrastructure that supports large scale distributed visualization is also introduced. Some typical visualization systems on Grids are referenced for discussions.     

Gallop: The Benefits of Wide-Area Computing for Parallel Processing

Internet computing has emerged, as an attractive paradigm for many applications that require access to distributed resources such as telemedicine, collaboratories, and transaction systems. For applications that require high performance, the Internet is assumed to be inappropriate. However, it is possible to obtain high performance for parallel scientific applications in an Internet environment. A wide-area scheduling system called Gallop has been developed to exploit opportunities for high performance using remote Internet resources for these applications. This paper describes the Gallop architecture and scheduling model, and performance results obtained for three parallel applications. The initial results indicate that wide-area parallel processing can lead to better performance even with Internet technology, but that current Internet bandwidth is a major bottleneck for file and binary transfer. This overhead limits the class of suitable applications to those that are large-grained or that will be run multiple times to amortize the transfer cost.     

Interactive view-dependent rendering over networks

For a client-server based view-dependent rendering system, the overhead of view-dependent rendering and the network latency are major obstacles in achieving interactivity. In this paper, we first present a multiresolution hierarchy traversal management strategy to control the overhead of view-dependent rendering for low-capacity clients. Then we propose a predictive parallel strategy to overcome the network latency for client-server based view-dependent multiresolution rendering systems. Our solution is to make the client process and the server process run in parallel, using the rendering time to cover the network latency. For networks with long round-trip times, we manage to overlap the network latency for one frame with the rendering time for multiple frames. View-parameters prediction is incorporated to make the parallelism of the client and the server feasible. In order to maintain an acceptable view-dependent rendering quality in the network environment, we develop a synchronization mechanism and a dynamic adjustment mechanism to handle the transient network slowdowns and the changes of the network condition. Our experimental results, in comparison with the sequential method, show that our predictive parallel approach can achieve an interactive frame rate while keeping an acceptable rendering quality for large triangle models over networks with relatively long round-trip times.     

ChattyGraph:面向异构多协处理器的高可扩展图计算系统

现阶段,随着数据规模扩大化和结构多样化的趋势日益凸现,如何利用现代链路内链的异构多协处理器为大规模数据处理提供实时、可靠的并行运行时环境,已经成为高性能以及数据库领域的研究热点.利用多协处理器(GPU)设备的现代服务器(multi-GPU server)硬件架构环境,已经成为分析大规模、非规则性图数据的首选高性能平台.现有研究工作基于Multi-GPU服务器架构设计的图计算系统和算法(如广度优先遍历和最短路径算法),整体性能已显著优于多核CPU计算环境.然而,这类图计算系统中,多GPU协处理器间的图分块数据传输性能受限于PCI-E总线带宽和局部延迟,导致通过增加GPU设备数量无法达到整体系统性能的类线性增长趋势,甚至会出现严重的时延抖动,进而已无法满足大规模图并行计算系统的高可扩展性要求.经过一系列基准实验验证发现,现有系统存在如下两类缺陷:(1)现代GPU设备间数据通路的硬件架构发展日益更新(如NVLink-V1,NVLink-V2),其链路带宽和延迟得到大幅改进,然而现有系统受限于PCI-E总线进行数据分块通信,无法充分利用现代GPU链路资源(包括链路拓扑、连通性和路由);(2)在...