DoubleP:一个有效的节点关键性能因素发现与聚类方法

随着处理器设计复杂性的提高,影响处理器节点性能的因素也越来越多.受资源(时间和空间)的限制,很难采集并分析所有影响性能因素的应用特征参数和机器性能轮廓参数.如何在众多的性能参数中提取影响性能的关键因素,成为并行系统节点性能评价中首先要解决的问题.提出了一个有效的DoubleP性能因素发现与聚类方法,该方法将众多复杂的性能因素聚焦到几个性能主成份上,明确了分析对象.通过DoubleP方法的分析,发现了14个影响处理器节点性能的关键因素和4个性能主成份.     

B/S架构应用软件性能测试问题研究

为了准确定位影响"浏览器/服务器(Brow ser/Server,B/S)"架构应用软件性能测试结果的主要因素,有效解决软件性能测试中遇到的问题,结合具体实践,全面分析了服务器网络性能、数据库调优能力、应用服务器参数设置、应用软件设计、服务器硬件等因素对B/S架构应用软件性能测试的影响,并提出了排除这些因素影响的方法。项目实践表明,这些方法较大程度地加快了应用软件性能测试的进程、提高了软件性能评估的准确性。     

CDMA无线蜂窝网中TCP性能分析及改善方法

在无线传输环境中,影响网络TCP性能的因素要比在有线传输中复杂的多,本文根据CDMA无线蜂窝网的特点,分析了其不同于传统有线通信网的影响网络TCP性能的因素,并根据这些因素,从物理层、链路层和传输层的角度给出了多种改善CDMA无线蜂窝网TCP性能的方法。     

计算机系统中通信机制优化分析

高性能并行计算机系统中由于CPU速度不断提高,通信机制成了影响系统性能的首要因素,为了优化通信性能,本文从硬件和软件两方面综合分析了影响通信的因素.     

基于正交试验的运动目标跟踪算法性能评价

针对目前运动目标跟踪算法性能评价中测试数据量大、试验次数多以及未充分考虑多因素组合场景下的算法性能表现等问题,提出一种基于正交试验的运动目标跟踪算法性能评价方法。分析影响算法性能的因素和水平,构建正交试验数据集,通过该数据集测试算法性能并利用极差分析法分析数据结果,以得到各影响因素间的强弱关系以及算法性能表现较好时的因素水平组合方式。分析结果表明,该方法能够全面、有效地评估运动目标跟踪算法的性能,减少测试次数和数据量,并为其他图像处理算法的性能评估提供参考。     

基于EPA的工业自动化网络延迟分析及改善

设法减少各种延迟是提高工业控制网络实时性的主要途径.深入讨论了EPA网络的传输通信以及节点处理的延迟,采用过程分析法研究了影响EPA网络响应时间的各种因素,探讨了不同因素对基于EPA的工业控制网络实时性能的影响程度,比较得出影响EPA实时性能的主要因子.通过对延迟影响因素的可控和不可控性能的分析,给出了控制和改善EPA实时性能的有效方法.     

压缩机用接线柱绝缘性能分析

绝缘不良是压缩机市场故障中危害性较大的一种,不仅可能导致压缩机报废,严重时还会危及人身安全。接线柱是影响压缩机绝缘性能的重要电气元件,而影响接线柱绝缘性能的因素有很多,如绝缘材料、温度、湿度、污损及试验电压等。一些专家曾先后发表有关压缩机接线柱的专利,从结构、材质、工艺、表面处理等方面来提高接线柱的绝缘性能,但针对接线柱绝缘性能影响因素的文献确鲜有报道。本文通过大量的试验研究,分析了影响压缩机接线柱绝缘性能的主要因素并提出改进措施。     

复杂电磁环境下信息化武器系统性能评估方法

根据信息作战原理,运用仿真和统计学分析的混合方法,建立基于概率的评估模型,在复杂电磁环境下,评估信息因素和火力因素对信息化武器系统性能的影响,对一个典型实例的初步研究表明:信息与火力联合打击能力是评估信息化武器系统性能的重要指标,信息因素与火力因素对联合打击能力的影响与该系统在复杂电磁环境下的工作能力有关,能力越弱则火力因素的影响就越强,能力越强则信息凶素的影响就越强,信息因素通常比火力因素更为重要.     

计算机网络性能分析与研究

计算机网络性能关系到网络的使用效率和成本,通过对影响计算机网络性能的系统潜在的因素进行分析研究,指出了协议和操作系统是影响计算机网络性能的主要因素,分析提出了优化系统性能设计的一些经验性规则,并研究总结出了一系列优化整个网络性能的措施。这些规则和措施的实施,必将会极大地优化网络的带宽效率,降低网络的运营成本。     

霍尔传感器磁路结构仿真分析与优化

针对特殊应用环境的霍尔电流传感器叠片铁芯磁路结构对传感器基本性能的影响进行分析.通过电磁场的基本理论,应用安培环路定律以及基尔霍夫磁路定律,推导出影响磁路性能的方程.结合Maxwel仿真工具,对于现有磁路结构进行仿真分析.提出环路截面不均匀、叠片间隙等影响磁路性能的因素,通过对其进行优化设计,改善各个影响因素.将优化后的磁路应用到传感器样机,使传感器的精度提升至0.04％,零点性能更加稳定.     

并行程序设计模型和语言

并行计算技术的发展已有20多年的历史了.时至今日,高性能并行计算仍然缺乏有效的并行程序设计方法和工具,使得编写并行程序、理解并行程序的行为、调试和优化并行程序的性能都很困难.从分析并行程序设计困难的原因入手,指出了当前各种高性能并行机系统支持的并行程序设计方法存在的诸多问题,综述了并行程序设计模型和语言的研究现状,给出了并行程序设计模型的评价标准,并提出了这一研究领域所面临的挑战性问题,指出了一些未来可能的发展方向.     

基于Docker的MPI和OpenMP混合编程

针对当前搭建集群并行系统复杂且耗时等问题,提出基于Docker搭建并行系统。介绍轻量级虚拟化技术Docker的核心概念和基本架构,并基于Docker技术在Linux平台上搭建集群并行开发环境。简要阐述并行计算的思想,叙述MPI和OpenMP并行计算的基本概念和特点,针对矩阵并行乘法的算法建立MPI和OpenMP的混合编程模型,并给出混合编程模型与MPI并行编程模型以及OpenMP并行编程模型的性能对比,分析出现差异的原因。基于该混合编程模型比较Docker与传统物理机两者搭建的并行系统的并行效率。     

面向国产异构众核系统的Parallel C语言设计与实现

异构众核架构具有超高的性能功耗比,已成为超级计算机体系结构的重要发展方向.但众核系统更为复杂的并行层次和存储层次,给编程和优化带来了极大的挑战,因此研究面向众核系统的并行编程技术,对于降低国产众核系统并行应用的编程难度、提升并行程序的性能都具有重要的意义.提出统一架构的多模式并行编程模型,包括异构融合的加速运算模型和按同构方式编程的自主运算模型,根据编程模型设计了Parallel C语言,能有效描述国产众核系统的异构并行性,与其它众核系统上MPI+X的使用模式相比,编程和系统优化都具有全局视角,在多级局部性描述、单边消息、兼容已有多核应用等方面具有特色;基于Open64构建了Parallel C编译系统,全面支持加速运算模型和自主运算模型,提出并实现了数据布局与自动DMA、编译指导的线程代理和拓扑位置感知的集合通信等优化.Micro Benchmark和实际应用在神威太湖之光计算机系统上的测试数据表明,Parallel C语言和编译系统具有良好的性能和可扩展性,能够有效支撑大型应用.     

MODELS AND TRENDS IN PARALLEL PROGRAMMING

This paper introduces and discusses programming models for parallel processing and recent trends in the area of parallel programming. The paper discusses different parallel programming languages and tools that reflect various parallel computation models. These language differ in expressiveness, portability and performance. Software design and implementation largely varies by using different languages that make the programmer task easy or complex. We describe here the design goals and the main issues of parallel programming models and languages belonging to the following categories: shared-space based languages, message-based languages, parallel toolkits, data-parallel languages, parallel declarative languages, parallel object-oriented languages, and parallel composition-based languages. Tools and languages such as HPF, Linda, Java, OpenMP, PVM, MPI, Parallel C+ +, Sisal, Orca, Mentat, SkieCL, BSP and others are described in some detail. Their main features for design and implementation of high performance applications are discussed. Finally, we outline directions of research and development in the parallel programming area with a special attention to novel approaches based on high-level programming structures that make transparent to the users the architectural details of parallel computing machines.     

一种支持容错的任务并行程序设计模型

任务并行程序设计模型已成为并行程序设计的主流,其通过发掘任务并行性来提高并行计算机的系统性能.提出一种支持容错的任务并行程序设计模型,将容错技术融入到任务并行程序设计模型中,在保证性能的同时提高系统可靠性.该模型以任务为调度、执行、错误检测与恢复的基本单位,在应用级实现容错支持.采用一种Buffer-Commit计算模型支持瞬时错误的检测与恢复;采用应用级无盘检查点实现节点故障类型永久错误的恢复;采用一种支持容错的工作窃取任务调度策略获得动态负载均衡.实验结果表明,该模型以较低的性能开销提供了对硬件错误的容错支持.     

Heterogeneous parallel and distributed computing

Heterogeneous network-based distributed and parallel computing is gaining increasing acceptance as an alternative or complementary paradigm to multiprocessor-based parallel processing as well as to conventional supercomputing. While algorithmic and programming aspects of heterogeneous concurrent computing are similar to their parallel processing counterparts, system issues, partitioning and scheduling, and performance aspects are significantly different. In this paper, we discuss the evolution of heterogeneous concurrent computing, in the context of the parallel virtual machine (PVM) system, a widely adopted software system for network computing. In particular, we highlight the system level infrastructures that are required, aspects of parallel algorithm development that most affect performance, system capabilities and limitations, and tools and methodologies for effective computing in heterogeneous networked environments. We also present recent developments and experiences in the PVM project, and comment on ongoing and future work.     

基于“魔方”的不同编程模型下科学应用性能比较

本文基于魔方(曙光5000A)"超并行"系统架构比较了不同并行编程模型下NAS Parallel Benchmark Multi-Zone的性能。结果表明,在单节点共享内存存储架构下OpenMP编程模型能获得更好性能,跨节点采用混合编程模型能获得更好性能,测试结果为百万亿次系统的使用及应用性能的提升起到指导作用。     

Overview of the KOAN programming environment for the iPSC/2 and performance evaluation of the BECAUSE test program 2.51

In this paper, we describe a Fortran programming environment using the KOAN Shared Virtual Memory on a distributed memory parallel architecture. We then discuss its use for parallelizing a benchmark application and present performance results. Initial findings show that it is possible to obtain good performance without the need for low level parallel programming.     

Data Parallel Algorithmic Skeletons with Accelerator Support

Hardware accelerators such as GPUs or Intel Xeon Phi comprise hundreds or thousands of cores on a single chip and promise to deliver high performance. They are widely used to boost the performance of highly parallel applications. However, because of their diverging architectures programmers are facing diverging programming paradigms. Programmers also have to deal with low-level concepts of parallel programming that make it a cumbersome task. In order to assist programmers in developing parallel applications Algorithmic Skeletons have been proposed. They encapsulate well-defined, frequently recurring parallel programming patterns, thereby shielding programmers from low-level aspects of parallel programming. The main contribution of this paper is a comparison of two skeleton library implementations, one in C++ and one in Java, in terms of library design and programmability. Besides, on the basis of four benchmark applications we evaluate the performance of the presented implementations on two test systems, a GPU cluster and a Xeon Phi system. The two implementations achieve comparable performance with a slight advantage for the C++ implementation. Xeon Phi performance ranges between CPU and GPU performance.     

Scalability of hybrid programming for a CFD code on the Earth Simulator

The Earth Simulator (ES) is an SMP cluster system. There are two types of parallel programming models available on the ES. One is a flat programming model, in which a parallel program is implemented by MPI interfaces only, both within an SMP node and among nodes. The other is a hybrid programming model, in which a parallel program is written by using thread programming within an SMP node and MPI programming among nodes simultaneously. It is generally known that it is difficult to obtain the same high level of performance using the hybrid programming model as can be achieved with the flat programming model.

In this paper, we have evaluated scalability of the code for direct numerical simulation of the Navier–Stokes equations on the ES. The hybrid programming model achieves the sustained performance of 346.9 Gflop/s, while the flat programming model achieves 296.4 Gflop/s with 16 PNs of the ES for a DNS problem size of 256³. For small scale problems, however, the hybrid programming model is not as efficient because of microtasking overhead. It is shown that there is an advantage for the hybrid programming model on the ES for the larger size problems.