异构分布式系统中一种新型主副版本调度算法

针对异构分布式系统中处理器数量相对较少时优先级约束条件带来的副版本调度易失败问题，提出一种新型高可靠性主副版本调度算法（HRPB）。任务模型以有向无环图（DAG）表示，该算法共计调度主、副两个版本的任务。在任务优先级排序阶段，根据任务执行时间及截止时限来制定新指标平均最晚开始时间（ALST）进行排序；在任务处理器分配阶段，采取多一重备份策略以解决处理器数量相对较少时优先级约束条件带来的副版本调度易失败问题，并且改进了副版本调度时的可靠性指标计算方法。通过随机生成DAG图进行算法仿真测试，实验结果表明，HRPB比eFRD具有更优的副版本调度成功率、更高的系统可靠性。     

Heterogeneous oxidation of indoor surfaces by gas‐phase hydroxyl radicals

We investigate heterogeneous oxidation kinetics of monolayer‐thick, surface‐sorbed organics, namely di‐n‐octyl phthalate (DnOP) and palmitic acid (PA), with gas‐phase OH. The pseudo‐first order rate constants for organic loss at OH concentrations of 1.6 × 10⁸ molecules/cm³ are: (2.3 ± 0.1) × 10^?4 to (4.8 ± 0.8) × 10^?4 s^?1, and (1.3 ± 0.5) × 10^?4 s^?1 for DnOP and PA, respectively. Films developed in indoor office environments over a few weeks are also oxidized using the same OH concentration. Heterogeneous decay rate constants of mass signals from these films, attributed to phthalates (MW = 390.6) and to PA, are similar to those for the single‐component films, ie, (1.9 ± 0.4) × 10^?4 to (3.4 ± 0.5) × 10^?4 s^?1, and (1.1 ± 0.4) × 10^?4 s^?1, respectively. These results suggest that the lifetimes for OH heterogeneous oxidation of monolayer‐thick indoor organic films will be on the timescale of weeks to months. To support this argument, we present the first analysis of the mass transfer processes that occur when short‐lived gas‐phase molecules, such as OH, are taken up by reactive indoor surfaces. Due to rapid chemical production, the diffusion limitation to mass transfer is less important for short‐lived molecules than for molecules with little chemical production, such as ozone.     

High-speed,two-dimensional digital image correlation algorithm using heterogeneous (CPU-GPU) framework

Two-dimensional digital image correlation (2D-DIC) is an experimental technique used to measure in-plane displacement of a test specimen. Real-time measurement of full-field displacement data is challenging due to enormous computational load of the algorithm. In order to improve the computational speed, the focus of recent research works has been on the approach of parallelization across subsets within image pairs using graphics processing unit (GPU). But alternate GPU-based parallelization approaches to improve the performance of this algorithm as per the order of data processing have not been explored. To address this research gap, our method utilizes parallelism within a subset as well as across subsets for each computation step in an iteration cycle. A heterogeneous (CPU-GPU) framework in combination with a pyramid-based initial values estimation for subsets (in parallel) is proposed in this work. The precompute steps of the proposed framework are implemented using CPU, whereas the main iterative steps are realized using GPU. It is demonstrated that the overall computational speed of the proposed heterogeneous framework improves by compared to a sequential CPU-based implementation for a pair of gray-scale images with a resolution of pixels. As an important milestone, feasibility to measure deformations in real time ( 1 s) is manifested in this study.     

Computational continua for linear elastic heterogeneous solids on unstructured finite element meshes

The computational continua framework, which is a variant of higher‐order computational homogenization theories that is free of scale separation, does not require higher‐order finite element continuity, and is free of higher‐order boundary conditions, has been generalized to unstructured meshes. The salient features of the proposed generalization are (i) a nonlocal quadrature scheme for distorted elements that accounts for unit cell distortion in the parent element domain and (ii) an approximate variant of the nonlocal quadrature that eliminates the cost of computing positions of the quadrature points in the preprocessing stage. The performance of the computational continua framework on unstructured meshes has been compared to the first‐order homogenization theory and the direct numerical simulation.     

小议经济型异构播出系统

电视播出系统的稳定性与安全性直接关系到电视播出的效果。结合实践经验,对经济型异构播出系统的构建进行介绍。     

Multi-Scale Analysis of Fretting Fatigue in Heterogeneous Materials Using Computational Homogenization

This paper deals with modeling of the phenomenon of fretting fatigue in heterogeneous materials using the multi-scale computational homogenization technique and finite element analysis (FEA). The heterogeneous material for the specimens consists of a single hole model (25% void/cell, 16% void/cell and 10% void/cell) and a four-hole model (25% void/cell). Using a representative volume element (RVE), we try to produce the equivalent homogenized properties and work on a homogeneous specimen for the study of fretting fatigue. Next, the fretting fatigue contact problem is performed for 3 new cases of models that consist of a homogeneous and a heterogeneous part (single hole cell) in the contact area. The aim is to analyze the normal and shear stresses of these models and compare them with the results of the corresponding heterogeneous models based on the Direct Numerical Simulation (DNS) method. Finally, by comparing the computational time and % deviations, we draw conclusions about the reliability and effectiveness of the proposed method.     

中高碳铝镇静钢形成MnS塑性夹杂的工艺开发与实践

 为提高中高碳钢产品的抗疲劳性能,利用中高碳钢的成分特点,研究开发了中高碳铝镇静钢中MnS以Al₂O₃为形核质点的非均质形核工艺,将钢中Al₂O₃脆性夹杂用塑性MnS包裹,解决了疲劳应力钢因脆性非金属夹杂引起的疲劳断裂问题。通过对微细、弥散Al₂O₃夹杂生成条件、MnS非均质形核析出热力学条件的研究,开展了钢中关键元素的成分设计、精炼及连铸集成工艺的设计与开发。工业实践表明,低活度氧条件下进行铝终脱氧可以形成3～5 μm微细弥散的Al₂O₃夹杂,并作为非均质形核的核心在二次枝晶晶间的凝固末端析出弥散、细小的粒状MnS;通过梯度脱氧、真空碳脱氧以及保护浇铸等操作可以有效稳定控制钢中全氧含量,提高钢水洁净度,成品T[O]质量分数平均为0.000 618%,较原工艺的0.000 739%降低了16%;成品的夹杂物中MnS及MnS包裹Al₂O₃夹杂所占比例大于96%,与世界领先产品的夹杂物控制水平相当,考虑到产品使用过程中Al₂O₃夹杂外部的MnS包裹层必须足够厚,塑性夹杂才能起作用,建立了MnS “有效包裹率” 的概念,当硬相夹杂物被MnS包裹且硬相夹杂物的最大半径不大于MnS包裹部位半径的1/2时,认为MnS对硬相夹杂物实现了“有效包裹”;MnS塑性夹杂工艺可明显提高材料的疲劳性能,成品的平均断裂韧性为83.47 MPa·m1/2,较原工艺的67.31 MPa·m1/2提高了24%。     

SMDP-based sleep policy for base stations in heterogeneous cellular networks

A dense heterogeneous cellular network can effectively increase the system capacity and enhance the network coverage. It is a key technology for the new generation of the mobile communication system. The dense deployment of small base stations not only improves the quality of network service, but also brings about a significant increase in network energy consumption. This paper mainly studies the energy efficiency optimization of the Macro-Femto heterogeneous cellular network. Considering the dynamic random changes of the access users in the network, the sleep process of the Femto Base Stations (FBSs) is modeled as a Semi-Markov Decision Process (SMDP) model in order to save the network energy consumption. And further, this paper gives the dynamic sleep algorithm of the FBS based on the value iteration. The simulation results show that the proposed SMDP-based adaptive sleep strategy of the FBS can effectively reduce the network energy consumption.     

格点QCD基础求解器及其异构计算实现的性能优化

格点量子色动力学(格点QCD)是研究夸克、胶子等微观粒子间相互作用的重要理论和方法. 通过将时空离散化为四维结构网格, 并将量子色动力学的基本场量定义在网格上, 让研究人员可以使用数值模拟方法, 从第一性原理出发研究强子间相互作用和性质, 但这个过程中的计算量极大, 需要进行大规模并行计算. 格点QCD计算的核心基础为格点QCD求解器, 是程序运行主要的计算热点模块. 本文研究在国产异构计算平台下格点QCD求解器的实现与优化, 提出一套格点QCD求解器的设计实现, 实现了BiCGSTAB求解器, 显著降低了迭代次数; 通过对奇偶预处理技术, 降低了所求问题的计算规模; 针对国产异构加速卡的特点, 优化了Dslash模块的访存操作. 实验测试表明, 相比优化前的求解器获得了约30倍的加速比, 为国产异构超算下格点QCD软件性能优化提供了有益的参考价值.     

基于5G边缘云的港口统一通信方案研究与实践

5G边缘云具有分布式、灵活组网、快速部署和本地流量分流等优势，在港口等垂直工业领域有很广泛的应用前景，同时对实现港口统一通信也提出了一定的要求。在研究基于5G边缘云的港口统一通信架构的基础上，提出了针对港口现场全域各类要素实现全面感知、支撑自动化业务改造的5G边缘云功能框架。此功能框架能很好地适用于港口场景的有线无线网络融合的统一承载，可作为目前复杂环境下的港口网络统一通信的网络基础设施。最后在某港口码头上对该方案进行了实际验证。