浅谈SMP机群上的并行计算

随着计算机硬件技术的高速发展,处理器和网络的性能不断地迅速提高和价格的日益下降,使得并行计算从传统的超级计算平台转移到由一组高性能节点或工作站/PC机构成的称之为机群的计算平台上,机群成为构建可扩放并行计算机的一大趋势。将主要从SMP机群的特点出发,分析适合SMP机群的并行计算模型、并行编程模型,并提出用MPI与OpenMP相配合来求解实际问题。     

基于EBP模型的服务化制造业集群知识共享风险评价

通过与传统制造业集群的对比,分析了服务化制造产业集群的知识共享过程,识别了服务化制造业集群知识共享过程中关系、利益与知识共享能力三方面的知识流并生风险,构建了基于EBP（熵值法和BP神经网络法）的服务化制造业集群知识共享风险组合评价模型。通过实证分析验证了模型的有效性与合理性,并通过与BP神经网络模型的误差对比分析,证实了该模型在知识共享风险识别方面的准确度更高。所提供的模型方法可为服务化制造业集群企业知识共享风险管理提供决策支持。     

隐藏式的设计方法在汽车内饰上的KANO模型研究

目的研究隐藏式的设计方法在汽车内饰应用上的优缺点,探讨隐藏式的设计在今后汽车内饰设计应用上的可行性。方法通过分析隐藏式设计的发展过程和优缺点,了解和把握隐藏式的设计方法在汽车内饰中的应用现状。归纳隐藏式的设计方法的造型方式,通过对比传统设计方法总结出隐藏式的设计方法的优缺点。并通过KANO模型调研隐藏式设计与传统设计在不同功能内的满意度,计算得到better-worse散点图,分析对比两者之间的差异,验证隐藏式设计的可行性。结果通过KANO模型的调研分析并引入电动汽车内饰设计实践,验证了隐藏式的设计可以提升汽车内饰的美观,并且满足现有车型内饰的功能需求。结论隐藏式的设计方法可以满足功能及美观的需要,提高内饰设计的自由度,并帮助汽车内饰造型创新是今后内饰设计发展的趋势之一。     

基于OpenMP的抛物方程(PE)声场并行计算方法

水声学理论和技术的发展对声场计算提出了越来越高的要求,如何充分利用高效的声场计算模型和现代计算机技术,实现声场的快速计算,已经成为现代水声技术的一个重要发展方向。多核PC机的出现和流行为提高声场计算的效率提供了一种有效的途径。在共享存储（OpenMP）模型的基础上,实现了对抛物方程计算模型RAM的并行计算编程。测试结果表明,利用该方法能够获得较高的并行计算效率。     

Development of Ubiquitous Simulation Service Structure Based on High Performance Computing Technologies

The simulation field became essential in designing or developing new casting products and in improving manufacturing processes within limited time, because it can help us to simulate the nature of processing, so that developers can make ideal casting designs. To take the prior occupation at commercial simulation market, so many development groups in the world are doing their every effort. They already reported successful stories in manufacturing fields by developing and providing the high performance simulation technologies for multipurpose. But they all run at powerful desk-side computers by well-trained experts mainly, so that it is hard to diffuse the scientific designing concept to newcomers in casting field. To overcome upcoming problems in scientific casting designs, we utilized information technologies and full-matured hardware backbones to spread out the effective and scientific casting design mind, and they all were integrated into Simulation Portal on the web. It professes scientific casting design on the NET including ubiquitous access way represented by ＂Anyone, Anytime, Anywhere＂ concept for casting designs.     

Fast hydrological model calibration based on the heterogeneous parallel computing accelerated shuffled complex evolution method

Hydrological model calibration has been a hot issue for decades. The shuffled complex evolution method developed at the University of Arizona (SCE-UA) has been proved to be an effective and robust optimization approach. However, its computational efficiency deteriorates significantly when the amount of hydrometeorological data increases. In recent years, the rise of heterogeneous parallel computing has brought hope for the acceleration of hydrological model calibration. This study proposed a parallel SCE-UA method and applied it to the calibration of a watershed rainfall–runoff model, the Xinanjiang model. The parallel method was implemented on heterogeneous computing systems using OpenMP and CUDA. Performance testing and sensitivity analysis were carried out to verify its correctness and efficiency. Comparison results indicated that heterogeneous parallel computing-accelerated SCE-UA converged much more quickly than the original serial version and possessed satisfactory accuracy and stability for the task of fast hydrological model calibration.     

Parallel Line Search in Method of Feasible Directions

In this paper the line search procedure within the method of feasible directions is parallelized and used in the solution of constrained structural optimization problems. As the objective function is linear in the variables, the step size problem reduces to a zero finding problem. That is, the step size is the distance along the direction vector to the nearest constraint boundary. Zero finding is accomplished in two steps—a march along the direction vector to bracket the zero followed by an interval reduction scheme. Both these steps are parallelized using MPI for message passing. When implemented on a cluster of workstations, for a convergence parameter of 10^–6, the time for optimization of composite pressure vessel reduces from 3 hours to hour when 64 processors are utilized, with a speedup of 7.0.     

低频噪声有调度计算模型分析

对于低频噪声,信号中心频率附近声压级较高的有调音部分发生变化时,包括谱峰的声压级增高和谱峰的形状变尖,会使主观有调感明显提高。因此在考虑有调度计算模型时,一定要充分考虑峰值信号部分产生的影响。现有的有调度计算模型虽然在此部分有所考虑,但是由于模型中的带宽计权因子不能恰当描述这一特征,因而不能准确描述相应的主观感觉,并且计算模型不稳定。通过实验研究与分析得到影响主观有调感最大的因素是有调音部分的声压级大小和谱峰形状,据此进一步从纯音的窄带频率调制信号的主观评价实验出发,通过实验所得的有调度评分值反演出有调度的带宽计权因子,从而对现有模型中的带宽计权因子进行了修正,增加了峰值尖锐程度对有调度的影响,提高了有调度模型的准确性,而且增强了计算的鲁棒性。     

并联构型装备的研究进展及若干关键技术

并联构型装备正处于商品化、产业化的关键时期。文章从结构、应用领域、驱动方式、控制等方面阐述了并联构型装备的研究进展，介绍了国际上的一些主要研究计划，分析了产业化进程中的若干关键技术及其可能的解决方案。     

平行板反应离子刻蚀系统均匀磁场的数值计算

在平行板反应离子刻蚀系统的圆筒形腔体的侧壁上绕上电流线圈后,系统中的磁场强度和均匀性都可进行调节。本文根据螺线管线圈内部磁场的分布规律,通过数值计算的方法在该系统中获得均匀的磁场。     

Clarification of an improved cluster renewal model (ICRM) for heat transfer characteristics in a rolling circulating fluidized bed (RCFB)

This paper presents a novel method to predict the gas layer thickness δ′ in an improved cluster renewal model (ICRM) to calculate the heat transfer coefficient h′_c for a rolling circulating fluidized bed (RCFB). Eulerian-Eulerian two-fluid model (TFM) with kinetic theory of granular flow is used to perform a numerical simulation. After comparing the pressure gradient −△p/△z′ and the simulated heat transfer coefficient between multiphase and the wall h_gs with previously published experiment data, the correctness and reliability of the simulation are able to be verified. In conclusion, firstly, the variation of h′_c calculated from δ′ in the ICRM is almost the same with h_gs in the case that the RCFB undergoes the rolling motion. In order to quantitatively evaluate the proposed novel method, an error percentage α between h′_c calculated from δ′ and h_gs is 2.043% which is less than 5%. This certified that the novel method to predict δ′ in the ICRM has higher accuracy to calculate h′_c. Secondly, h′_c calculated from δ′ is mainly influenced by rolling amplitude Θ rather than by rolling period T. Specifically, with the increase of Θ, amplitudes of T_s are decreased, which is caused by the increased heat transfer and the decreased δ′ between cluster and wall. The decreased amplitudes of T_s are able to increase heat transfer efficiency, which eventually increases h′_c. Thirdly, in the case that normalized rolling period t/T is changed from 0 to 0.5, higher h′_c calculated from δ′ in Region Ⅰ results from the large local cross-sectional particle volume fraction c_local in Region Ⅰ with a smaller δ′, while lower h′_c calculated from δ′ in Region III results from the smaller c_local in Region III with a larger δ′. Therefore, it is concluded that h′_c calculated from δ′ is able to well predict the bed-to-wall heat transfer coefficient of the RCFB even if the rolling motion is considered.     

IMS实时计费系统集群的负载均衡研究

对IP多媒体子系统(IMS)中的实时计费系统(OCS)集群进行了研究,针对其高集群异构性和高请求负载异构性特征,提出了一种最小公倍数能力(LCMC)负载均衡算法.此算法考虑了各个服务器处理能力的差异和各个服务器当前的负载,并特别考虑了不同请求产生的负载差异,采用概率的方式进行请求分发.实验结果表明,在高集群异构性和高请求负载异构性的系统中,LCMC算法明显优于一些传统的负载均衡算法,能很好地满足IMS中OCS集群独有的特征.