HiCOO: Hierarchical cooperation for scalable communication in Global Address Space programming models on Cray XT systems

Global Address Space (GAS) programming models enable a convenient, shared-memory style addressing model. Typically this is realized through one-sided operations that can enable asynchronous communication and data movement. With the size of petascale systems reaching 10,000s of nodes and 100,000s of cores, the underlying runtime systems face critical challenges in (1) scalably managing resources (such as memory for communication buffers), and (2) gracefully handling unpredictable communication patterns and any associated contention. For any solution that addresses these resource scalability challenges, equally important is the need to maintain the performance of GAS programming models. In this paper, we describe a Hierarchical COOperation (HiCOO) architecture for scalable communication in GAS programming models. HiCOO formulates a cooperative communication architecture: with inter-node cooperation amongst multiple nodes (a.k.a multinode) and hierarchical cooperation among multinodes that are arranged in various virtual topologies. We have implemented HiCOO for a popular GAS runtime library, Aggregate Remote Memory Copy Interface (ARMCI). By extensively evaluating different virtual topologies in HiCOO in terms of their impact to memory scalability, network contention, and application performance, we identify MFCG as the most suitable virtual topology. The resulting HiCOO architecture is able to realize scalable resource management and achieve resilience to network contention, while at the same time maintaining or enhancing the performance of scientific applications. In one case, it reduces the total execution time of an NWChem application by 52%.     

无序体系薛定谔算子格林函数及谱分布

无序体系通常有着独特的光电性质,随着材料应用的深入,无序体系渐渐引起关注.研究无序体系的物理性质,最重要的便是它的电子性质,这与薛定谔算子谱的性质有关.而研究薛定谔算子谱的性质,关键是要研究格林函数的衰减性.它从数理角度描述了微观粒子的运动状态,对相应物理现象的解释和预测很有意义.通过两类典型无序体系薛定谔算子格林函数的衰减性质及谱分布的相关推论,可以看出,两种算子具有一定相似性,但目前对这两种算子格林函数衰减性的研究采用的是不同的方法.后续可以寻求更统一的对准周期体系和随机体系或其他无序体系特征函数局域化的研究方法,并推广到更一般的无序体系中.     

一种基于机器人的航天器大型部件自主装配方法

航天器大型部件在狭小空间中装配存在着视野受限、目标位置不可见等问题,传统的吊装方法无法精确调整位姿、磕碰风险极高．本文提出一种基于双目视觉定位的机器人辅助装配路径规划方法．该方法以航天器、待安装部件以及机器人等装配要素的3维模型为基础,采用双目视觉系统对航天器上安装位置的几何特征进行精确定位．利用测量结果确定机器人和航天器之间的相对位姿关系,进而构建航天器装配现场的虚拟环境,识别装配过程中的几何约束,然后利用轴对齐包围盒方法进行碰撞检测,采用随机路图法规划出一条无干涉的装配路径,最后利用离线编程技术生成机器人可执行的装配序列．以某型号航天器激光测高仪为对象开展装配试验,完成了大型部件在狭小凹舱内的无磕碰安装．结果表明该方法可以实现对螺纹孔精准定位,快速规划出无干涉装配路径,能够控制机器人安全、高效完成大型部件在狭小空间下的安装作业．     

弹性层中的SH板波传播特性及电磁超声激励方法的研究

研究水平剪切波(Shear Horizontal,简称SH波)的传播特性及激励接收方法对该波在无损检测和声弹性技术范围内的工程应用有着重要的价值.首先,SH波在弹性层中传播的多模态、频散特性在文中作了详细的分析,同时重点研究了电磁超声技术在非铁磁性材料波导中产生SH波的探头结构和基础理论,并以试验的方法验证了其部分电声特性,结果表明,SH板波具有其他类型波所没有的独特特性,而电磁超声技术在SH波的激励接收方面有着很大的优势.     

基于DM7520的在线漏磁信号采集系统

对一种基于工控机PC104和数据采集卡DM7520的在线漏磁信号采集系统进行了研究,并提出了相应的滤波算法和基于小波多分辨率分析的数据压缩算法。通过系统的实际应用,在现场试验中取得快速、精确、可靠的测试效果。     

关于电算化“甩帐率低”的解决办法

０引言现在企业中应用的财务软件不外乎两种情况：一种是商业化财务软件，另一种是结合单位实际情况开发的财务软件。二者在帐务处理上基本是按照现行财务制度要求设计的，但普遍存在“甩帐率低”的问题。究其原因主要有以下两方面：一是初始化工作不够规范，二是电算化记...     

Kilogram-scale synthesis of Mg-Al-CO3 LDHs nanosheets using a new two-stage reactor

A novel reactor with a two-stage reaction chamber was designed to kilogram-large-scale synthesized Mg-Al-CO₃ layered double hydroxide (Mg-Al-CO₃ LDHs) nanosheets. The effects of the reactant concentration and rotation speed of the dynamic structure on the crystal size and morphology were investigated. The results show that the size of Mg-Al-CO₃ LDH particles gained in typical condition is about 50 nm in length and 5.7 nm in thickness, and the size becomes larger as the reactant concentration or rotationl speed decreased. The possible reaction mechanism was explained by three critical factors of the reactor, N₂ gas and polyethylene glycol 2000.     

钢塑复合冲淤管的开发与应用

针对当前围堰造地与环境保护之间出现的问题,开发了一种钢塑复合冲淤管。经过工程试验证明这是一种切实可行的新技术,使用该产品做成的冲淤坝具有很好的冲淤能力。该产品的推广使用可减小围堰造地对环境的影响,加快冲淤速度,缩短围堰造地周期,节约造地成本。     

Nucleation Regulation and Anchoring of Halide Ions in All-Inorganic Perovskite Solar Cells Assisted by CuInSe2 Quantum Dots

Ionic feature of halide perovskites may lead to the formation of defect states in the polycrystalline films, which can deteriorate the device performance. To solve this issue, a low-temperature seed-assisted growth method can contribute to defect-passivated CsPbI_1.2Br_1.8 perovskite films by introducing CuInSe₂ quantum dots (QDs). As a result, the migration channels inside perovskite grains can be obviously suppressed, but few positive effects on the ions migration in grain boundaries. Conversely, this appearance demonstrate that the location of CuInSe₂ QDs is grain interior, but not grain boundary. Meanwhile, the lower defect density can help promoting device efficiency from 8.97% to 10.26%, which is among the high-efficiency level. Besides, this work can also provide an in-depth insight into the issue of photoinduced halide segregation.