大规模生物网络马尔可夫聚类的并行化算法

马尔可夫聚类算法（MCL）是在大规模生物网络中寻找模块的一个有效方法，能够挖掘网络结构和功能影响力较大的模块。算法涉及到大规模矩阵计算，因此复杂度可达立方阶次。针对复杂度高的问题，提出了基于消息传递接口（MPI）的并行化马尔可夫聚类算法以提高算法的计算性能。首先，生物网络转化成邻接矩阵；然后，根据算法的特性，按照矩阵的规模判断并重新生成新矩阵以处理非平方倍数矩阵的计算；其次，并行计算通过按块分配的方式能够有效地实现任意规模矩阵的运算；最后，循环并行计算直至收敛，得到网络聚类结果。通过模拟网络和真实生物网络数据集的实验结果表明，与全块集体式通信（FCC）并行方法相比，平均并行效率提升了10个百分点以上，因此可以将该优化算法应用在不同类型的大规模生物网络中。     

On the design of a new dynamic credit-based end-to-end flow control mechanism for HPC clusters

High Performance Computing usually leverages messaging libraries such as MPI, GASNet, or OpenSHMEM, among others, in order to exchange data among processes in large-scale clusters. Furthermore, these libraries make use of specialized low-level network layers in order to achieve as much performance as possible from hardware interconnects such as InfiniBand or 40 Gb Ethernet, for example. EXTOLL is an emerging network targeted at high performance clusters.Specialized low-level network layers require some kind of flow control in order to prevent buffer overflows at the receiver side. In this paper we present a new end-to-end flow control mechanism that is able to dynamically adapt, at execution time, the buffer resources used by a process according to the communication pattern of the parallel application and the varying activity among communicating peers. The tests carried out on a 64-node 1024-core EXTOLL cluster show that our new dynamic flow control mechanism presents very low overhead with an extraordinarily high buffer efficiency, as overall buffer resources are reduced by 4× with respect to the amount of buffers required by a static flow control protocol achieving similar low overhead levels.     

有限元SN中子输运模拟的区域分解并行

确定论中子输运方法具有计算速度快、可获取物理量的精细场分布、可高效多物理耦合等优点，随着有限元方法在中子输运模拟中的应用，复杂几何结构、大尺度下的屏蔽问题和临界问题都能得到高保真建模和分析。离散纵标（S_N）法是求解中子输运方程的有效数值方法，基于OpenMP并行机制对各独立离散方向进行并行求解，可提高S_N输运模拟的计算速度，但并行规模较有限。对几何空间进行区域分解并采用MPI并行机制，可实现大规模并行扩展，进而实现对大型问题的高精度快速求解。本文采用并行自适应非结构网格应用框架JAUMIN进行区域分解和进程间通信，通过并行S_N扫描实现了自主有限元输运程序ENTER的高效并行，完成正确性检验后在天河Ⅱ号超级计算机上使用1 440个CPU核完成了1.43×10⁷网格单元、2.81×10⁹自由度规模问题的测试，计算时间约7.4 h。表明该程序具备了有效模拟大型复杂结构中子输运问题的能力，具有一定工程应用价值。     

基于Windows平台的PC集群并行计算技术

介绍一种基于Windows系统平台的、用于实现高性能计算的PC机群并行环境,详细描述了它的体系结构及其构建中的关键技术;并结合一个实际系统的应用实例简单说明了其构建方法和在其环境中进行并行程序设计的基本过程。     

MPI在生物信息学中的应用

高性能计算是计算机科学的一个分支,研究并行算法和开发相关软件,致力于开发高性能计算机。在生物技术领域,高性能计算机已成为科学研究的必备工具。生物信息学(Bioinformatics)是现代生命科学与计算机科学、数理科学、化学等领域相互交叉而形成的一门新兴学科,实质就是利用计算机科学和网络技术来解决生物学问题,文章简单介绍了用于并行高性能计算的MPI接口,以及生物信息学中基于MPI库的一些相关软件工具。     

基于云平台的MPI研究与应用

为满足信息化处理需求的增长,提出MPI在云计算领域的应用研究。文章在介绍了云计算的优势、MPI概念的基础上,将MPI并行程序与并行机群系统及云计算平台整合在一起,借助Hadoop框架,充分利用MapReduce模式来实现并行化云计算对应的MPI算法。研究结果分析表明,该算法是可行的,可有效提高MPI并行处理的性能。     

Flow visualization of Rayleigh–Bénard convection for cubical cavity

Natural convective flow of air inside the cubical cavity is investigated numerically. The temperature of the bottom wall is kept higher than that of top cold wall, and other four walls are assumed to be adiabatic. Attention has been paid to the convective discretization schemes, like upwind, QUICK, total variation diminishing, normalized variable diagram (NVD) schemes that are compared with respect to accuracy. The output is validated with respect to the results available in the literature. A parallel computing message passing interface code is adapted to run the simulations. From the results, it is observed that the NVD scheme gives better results among all the employed convective discretization schemes irrespective of the mesh structure. Thus, in this article, self filtered central differencing which is a family of NVD, is used. From the enormous output data, along with the streamlines, contours of isotherms, the new technique of energy pathlines, and field synergy are used to visualize the fluid flow and heat transfer mechanism arising in the system in the range of Ra from 10³ to 10⁶. Free energy streamlines are observed with small Ra, whereas trapped energy streamlines are observed with high Ra. When Ra increases, synergy angle increases and implies that the synergy between the velocity vector and temperature gradient gets reduced and leads to increasing values of average Nusselt number (Nu).     

基于 MPI ＋OpenMP 的红外弱小目标检测并行计算

为有效监控红外弱小目标运动的全过程，必须采用多个波段同时探测，但是多波段探测必然带来计算时间的大幅增长，无法满足实际应用中对目标检测实时性的要求。针对这一问题，本文提出一种基于MPI＋OpenMP的层次化并行方法，充分利用消息传递模型和共享存储模型的优势，并基于多处理器节点集群进行测试。实验结果表明，该并行程序在保证相同的检测概率的情况下加速比达到8．61，极大地提高了目标检测的效率。     

Huge-scale molecular dynamics simulation of multibubble nuclei

We have developed molecular dynamics codes for a short-range interaction potential that adopt both the flat-MPI and MPI/OpenMP hybrid parallelizations on the basis of a full domain decomposition strategy. Benchmark simulations involving up to 38.4 billion Lennard-Jones particles were performed on Fujitsu PRIMEHPC FX10, consisting of 4800 SPARC64 IXfx 1.848 GHz processors, at the Information Technology Center of the University of Tokyo, and a performance of 193 teraflops was achieved, which corresponds to a 17.0% execution efficiency. Cavitation processes were also simulated on PRIMEHPC FX10 and SGI Altix ICE 8400EX at the Institute of Solid State Physics of the University of Tokyo, which involved 1.45 billion and 22.9 million particles, respectively. Ostwald-like ripening was observed after the multibubble nuclei. Our results demonstrate that direct simulations of multiscale phenomena involving phase transitions from the atomic scale are possible and that the molecular dynamics method is a promising method that can be applied to petascale computers.     

KNEM: A generic and scalable kernel-assisted intra-node MPI communication framework

The multiplication of cores in today’s architectures raises the importance of intra-node communication in modern clusters and their impact on the overall parallel application performance. Although several proposals focused on this issue in the past, there is still a need for a portable and hardware-independent solution that addresses the requirements of both point-to-point and collective MPIoperations inside shared-memory computing nodes.