三维特征线的并行方法研究

设计了基于角度并行的三维特征线的并行算法。同时,为提高并行效率,采用了角度预分组和射线组合的策略,使通信量达到最小,并应用到中子输运方程的三维特征线计算程序TCM中。数值结果表明：角度并行方法和串行方法的结果完全一致,在采用了通信优化和负载平衡策略后,并行效率得到了显著提高。     

Enhanced balancing Neumann–Neumann preconditioning in computational fluid and solid mechanics

In this work, we propose an enhanced implementation of balancing Neumann–Neumann (BNN) preconditioning together with a detailed numerical comparison against the balancing domain decomposition by constraints (BDDC) preconditioner. As model problems, we consider the Poisson and linear elasticity problems. On one hand, we propose a novel way to deal with singular matrices and pseudo‐inverses appearing in local solvers. It is based on a kernel identification strategy that allows us to efficiently compute the action of the pseudo‐inverse via local indefinite solvers. We further show how, identifying a minimum set of degrees of freedom to be fixed, an equivalent definite system can be solved instead, even in the elastic case. On the other hand, we propose a simple implementation of the algorithm that reduces the number of Dirichlet solvers to only one per iteration, leading to similar computational cost as additive methods. After these improvements of the BNN preconditioned conjugate gradient algorithm, we compare its performance against that of the BDDC preconditioners on a pair of large‐scale distributed‐memory platforms. The enhanced BNN method is a competitive preconditioner for three‐dimensional Poisson and elasticity problems and outperforms the BDDC method in many cases. Copyright © 2013 John Wiley & Sons, Ltd.     

Scheduling algorithms‐by‐blocks on small clusters

The arrival of multicore architectures has generated an interest in reformulating dense matrix computations as algorithms‐by‐blocks, where submatrices are units of data and computations with those blocks are units of computation. Rather than directly executing such an algorithm, a directed acyclic graph is generated at runtime that is then scheduled by a runtime system such as SuperMatrix. The benefit is a clear separation of concerns between the library and the heuristics for scheduling. In this paper, we show that this approach can be taken one step further using the same methodology and an ad hoc runtime to map algorithms‐by‐blocks to small clusters. With no change to the library code, and the application that uses it, the computational power of such small clusters can be utilized. An impressive performance on a number of small clusters is reported. As a proof of the flexibility of the solution, we report performance results on accelerated clusters based on graphics processors. We believe this to be a possible step towards programming many‐core architectures, as demonstrated by a port of the solution to Intel's Single‐chip Cloud Computer (Intel, Santa Clara, CA, USA). Copyright © 2012 John Wiley & Sons, Ltd.     

Analysis of spinning process using the Taguchi method. Part I: Effect of spinning process variables on fibre orientation and tenacities of sliver and roving

Abstract

The effect of drafts at different stages of spinning process on fibre orientation and tenacities of sliver and roving has been studied by using Taguchi method and analysis of variance. Among all the fibre orientation parameters analysed the ‘coefficient of relative fibre parallelization (K _rp)’ influenced the maximum with change in process variables. Further, the equation used for calculation of K _rp was critically analyzed and modified for a better explanation of fibre orientation. The fibre orientation in sliver and roving was also studied in terms of a new index of measurement of fibre straightness along the length. Increase in lap hank and card draft increases proportion of curved fibre ends (ρ) and decreases K _rpm in card and breaker sliver. However, the improvement in K _rpm is seen to be higher in finisher sliver and roving produced by finer lap hank and higher card draft in subsequent processes. Increase in lap hank, card draft and draft/doublings decrease the FSI in slivers. The lower speed frame draft tends to decrease fibre straightening index (FSI) of roving. Further, a combination of K _rpm, ρ and FSI values is seen to decide the strength of slivers.     

对称分支并行微通道中气液两相流的均匀性规律

采用高速摄像系统研究了对称分支形并行微通道内气液两相流及弹状气泡均匀性规律。实验中分别采用含0.3% SDS的甘油-水溶液与氮气作为液相和气相。观察到弹状流和泡状流两种流型,作出了由两相操作条件构成的流型图及流型转变线。结果表明,气泡非均匀性主要由两微通道内流体之间的相互作用、下游通道中流体动力学的反馈作用以及通道制造误差造成。随液相黏度增大,气泡均匀性变好;在高液相流量以及低气相压力下操作,气泡尺寸分布更易达到均匀。基于压力降守恒原理和微通道内气液两相流阻力模型,构建了两通道中气泡尺寸的预测模型。     

基于并行机制的商务智能系统BI-PaaS

针对海量信息处理对传统的商业智能软件提出的挑战,设计并实现了一种基于并行机制的商务智能(BI)系统(BI-PaaS)。该系统搭建在中国移动大云基础设施之上, 以云计算的强大并行计算和分布式存储能力为支撑,将ETL、DM、OLAP、Report等各类BI能力并行化。实验结果表明,基于并行机制的BI-PaaS系统可以大大提高BI能力的数据处理性能,有效地支持电信运营的海量数据分析。     

三维多群中子输运蒙特卡罗程序MCMG-Ⅱ基准检验

三维多群P₃中子输运蒙特卡罗程序MCMG通过版本更新和功能扩充，能够配备各种多群微观、宏观截面库，截面输入文件进一步简化，版本从Ⅰ发展到Ⅱ，参数更新到ENDF/B-Ⅶ库。程序发展了针对物质的碰撞机制，具有并行计算功能。对于12个临界基准问题和1个外源问题，MCMG-Ⅱ计算取得了与实验和连续截面MCNP 5程序一致的结果，计算速度较MCNP-5提高了3~6倍。     

新一代Linux启动技术的比较与测试

目前,Linux的使用越来越广泛,但Linux启动速度不尽如人意,尤其当Linux作为服务器使用时,如果出现故障,需要重起机器,就更强调其启动速度.此外,现在的Linux系统也更强调服务的可用性.从Linux传统的启动sysvinit出发,介绍下一代的Linux启动的替代程序发展过程与现状,并分析各个替代应用的普遍性,服务依赖关系解决的手段,服务的监管.此外,还用专门的启动分析程序bootchart分析了替代程序的各方面的启动性能,并预测其应用前景.     

基于MapReduce的朴素贝叶斯算法文本分类方法

为了解决传统串行朴素贝叶斯算法分类性能低下的问题,提出一种基于朴素贝叶斯算法的并行化分类方法。选取多项式朴素贝叶斯,搭建Hadoop集群,通过卡方检验选取特征词,利用词频-逆文本频率指数方法计算出每个特征项的权值,并求出每类的权重总和,将权值应用到朴素贝叶斯公式中得到分类结果。实验结果表明:在该集群上设计的并行化朴素贝叶斯分类方法较比传统朴素贝叶斯方法,其精确率,召回率,F1值分别至少提高了7.66%,7.56%,11.98%,且用时更短,说明本文的方法能够提高处理文本的时间效率。     

Multi-core optimization for conjugate gradient benchmark on heterogeneous processors

Developing parallel applications on heterogeneous processors is facing the challenges of ‘memory wall’, due to limited capacity of local storage, limited bandwidth and long latency for memory access. Aiming at this problem, a parallelization approach was proposed with six memory optimization schemes for CG, four schemes of them aiming at all kinds of sparse matrix-vector multiplication (SPMV) operation. Conducted on IBM QS20, the parallelization approach can reach up to 21 and 133 times speedups with size A and B, respectively, compared with single power processor element. Finally, the conclusion is drawn that the peak bandwidth of memory access on Cell BE can be obtained in SPMV, simple computation is more efficient on heterogeneous processors and loop-unrolling can hide local storage access latency while executing scalar operation on SIMD cores.