用于高性能计算的作业调度能效性研究综述

由于科学研究与商业应用等对高性能计算的需求与日俱增,高性能计算的性能和系统规模得到迅速发展。但是,急剧增长的功耗严重限制了高性能计算系统的设计和使用,使得低功耗技术成为高性能计算领域的关键技术。作为整个系统的核心组件,作业调度系统立足有限的系统资源,对用户提交的应用进行作业-资源分配,其能效性对于整个高性能计算系统的能耗控制与调节起到至关重要的作用。首先介绍主要的能量效率技术和常用的作业调度策略,然后对当前高性能计算作业调度能效性进行分析,并讨论了其面临的挑战及未来发展方向。     

企业能耗过程性质判定及能量守恒性分析

为实现工业企业能源消耗过程行为和结构特性的分析,在企业能耗过程模型的基础上,研究了与模型相关的动态性质与结构性质,并有针对性地分析了网系统的能量守恒性。为实现生产中各物质流的动态逻辑关系以及工艺流程静态拓扑结构的描述,提出了动态与静态关联矩阵的概念并分别给出定义;分别研究了与动态行为有关的动态平衡量、稳态运行区间等性质;在模型加权守恒性研究的基础上,提出了一种基于S不变量的网系统能量守恒性计算方法并延伸出对设备与产品能效的分析计算。分析表明,方法有助于实际能源消耗过程中物料平衡性、产品或原料积压以及产品能效等特性的分析与检查。     

The thermodynamics of several elements at high pressure

We have used the CALPHAD-compatible equation of state (EOS) based on the explicit Gibbs free energy concept for the solid state of ten important elements: V, Nb, Ta, Mo, W (groups VB and VIB), Pd, Pt (group VIIIB) and Cu, Ag and Au (group IB). The new formulation uses SGTE data for ambient pressure and converges to the quasi-harmonic model at the limit of extreme pressure to calculate the Gibbs free energy as a function of pressure and temperature. The model is based on the available pressure-volume-temperature (PVT) data on the elements and can be usefully extrapolated to extreme pressures. When compared to shock wave data, the modeled EOS holds well, but the fit is not totally satisfactory in the ultrahigh-pressure range. A great advantage of this formulation is that it can be used to calculate thermodynamic properties such as the heat capacity and entropy at very high temperatures and pressures.     

几种协作分集模式能量消耗比较

为了提高通信性能和延长无线网络生命周期,提出了统一的系统模型;分析了选择译码转发(SDF)和多输入多输出(MIMO)协作工作模式,研究了影响能耗的几个因素。研究表明,对于同一平均误比特率值,SDF系统比单输入单输出(SISO)系统的能量消耗低;当发射节点数目固定时,MIMO协作方式接收节点不是越多越好,而是存在一个最佳数目。短距离时,协作MIMO系统比传统SISO系统的比特能耗高。大于临界距离时,协作MIMO系统才能节省能量。     

高能物理事例级数据管理与传输系统的研究

高能物理实验不断的进步与发展产生了PB乃至EB级的数据,这些数据的采集、存储、传输与共享、分析与管理都面临着极大的问题与挑战。为了应对这些挑战,设计和实现了面向事例的数据管理系统,有效解决事例数据处理效率低以及分站点资源利用率低的问题。设计了一个基于Nosql数据库的事例索引系统,通过事例数据特征抽取,选取物理学家最感兴趣的属性作为索引,存储在数据库中,并采用倒排索引技术,提高事例数据检索的效率。针对事例数据进行缓存优化,减少数据转化和存储开销。提出数据跨域传输方案,充分利用网络带宽,降低分站点处理数据的延迟。系统进行了相关验证,实验结果表明,事例级的索引技术能够显著提高事例数据的检索效率,数据传输系统的网络带宽也可以利用到百分之九十以上。     

Lower bound of tracking performance with finite control energy and channel energy constraint

This paper studies the tracking performance of the single-input single-output (SISO), finite dimensional, linear and time-invariant (LTI) system over an additive white Gaussian noise (AWGN) channel with finite control energy and channel input energy constraint. A new performance index is proposed which is minimized over all stabilizing two-degree-of-freedom controllers. The explicit expressions of the lower bound of the tracking performance and the minimum of signal-to-noise required are obtained. The results show that the lower bound is correlated to the unstable pole, nonminimum phase zero and the channel scaling factor. Finally, one example is given to validate the conclusions by adopting the special inner-outer factorization.     

Computational determination of binding structures and free energies of glucose 6-phosphate dehydrogenase with novel steroid inhibitors

Glucose 6-phosphate dehydrogenase (G6PD), the first and the rate-limiting enzyme in the pentose phosphate pathway (PPP), catalyzes the oxidation of G6P to 6-phosphogluconolactone and the reduction of NADP⁺ to NADPH. Its key role in cancer promotes the development of a potent and selective inhibitor that might increase cancer cell death when combined with radiotherapy. In the present study, we investigated the detailed binding modes and binding free energies for G6PD interacting with a promising series of recently developed inhibitors, i.e., the steroid derivatives, by performing molecular docking, molecular dynamics (MD) simulations, and binding free energy calculations. The docking indicates that the inhibitors occupy the binding sites of both G6P and NADP⁺. The calculated binding free energies on the basis of the MD-simulated enzyme–inhibitor complexes are in good agreement with the experimental activity data for all of the examined inhibitors. The valuable insights into the detailed enzyme–inhibitor binding including the important intermolecular interactions, e.g., the hydrogen bond interaction and the hydrophobic interaction, have been provided. The computational results provide new insights into future rational design of more potent inhibitors of G6PD as a treatment for cancer.     

A PM7 dynamic residue-ligand interactions energy landscape of the BACE1 inhibitory pathway by hydroxyethylamine compounds. Part I: The flap closure process

BACE1 is an enzyme of scientific interest because it participates in the progression of Alzheimer’s disease. Hydroxyethylamines (HEAs) are a family of compounds which exhibit inhibitory activity toward BACE1 at a nanomolar level, favorable pharmacokinetic properties and oral bioavailability. The first step in the inhibition of BACE1 by HEAs consists of their entrance into the protease active site and the resultant conformational change in the protein, from Apo to closed form. These two conformations differ in the position of an antiparallel loop (called the flap) which covers the entrance to the catalytic site. For BACE1, closure of this flap is vital to its catalytic activity and to inhibition of the enzyme due to the new interactions thereby formed with the ligand. In the present study a dynamic energy landscape of residue-ligand interaction energies (ReLIE) measured for 112 amino acids in the BACE1 active site and its immediate vicinity during the closure of the flap induced by 8 HEAs of different inhibitory power is presented. A total of 6.272 million ReLIE calculations, based on the PM7 semiempirical method, provided a deep and quantitative view of the first step in the inhibition of the aspartyl protease. The information suggests that residues Asp93, Asp289, Thr292, Thr293, Asn294 and Arg296 are anchor points for the ligand, accounting for approximately 45% of the total protein-ligand interaction. Additionally, flap closure improved the BACE1-HEA interaction by around 25%. Furthermore, the inhibitory activity of HEAs could be related to the capacity of these ligands to form said anchor point interactions and maintain them over time: the lack of some of these anchor interactions delayed flap closure or impeded it completely, or even caused the flap to reopen. The methodology employed here could be used as a tool to evaluate future structural modifications which lead to improvements in the favorability and stability of BACE1-HEA ReLIEs, aiding in the design of better inhibitors.     

Binocular energy response based quality assessment of stereoscopic images

Perceptual quality assessment in three-dimensional (3D) is challenging. In this research, we propose a binocular energy response based quality assessment metric of stereoscopic images. To be more specific, we first construct binocular energy responses of the original and distorted images, and measure the similarity between them as Image Quality Metric (IQM). Then, the binocular response and the binocular masking components are used to modulate the IQM, respectively. Finally, two evaluation results are nonlinearly integrated into an overall score by considering the importance of each component. Experimental results show that compared with the relevant existing metrics, the proposed metric can achieve higher consistency with the subjective assessment of stereoscopic image.     

LNG冷能利用方案的本质安全性评价

为了获得本质安全性较高的LNG冷能利用方案,利用本质安全指标法评价了3个可选方案,根据评价结果对方案进行了优选。首先介绍了本质安全及典型的指标型本质安全评价方法,阐述了本质安全指标法的结构及其评分体系;应用本质安全指标法分别评价了LNG与空气换热,LNG与污氮换热,LNG与氮气、空气换热3个LNG冷能回收利用方案,得到它们的本质安全指标值分别为21,20,18;结果表明LNG与氮气、空气换热方案的本质安全指标值最低,即其本质安全性最高,主要原因是用氮气作中间介质能够消除潜在的LNG与空气、或污氮混合而引发爆炸的可能性;最后,确定该方案的本质安全性最佳,并给出了该方案的流程图。本研究改良了LNG冷能利用方案的本质安全性并对方案的优选提供了数据支持。     

Optimal tracking performance of communication-constrained systems under energy constraints

This paper investigates the optimal tracking performance of systems by considering packet dropouts, additive white Gaussian noise(AWGN), and coding under energy constraints. The optimal tracking performance of communication-constrained systems is obtained by spectral decomposition and partial factorization. The results show that the optimal tracking performance of the system is related to intrinsic properties such as non-minimum phase zeros and unstable poles. What's more, encoding, data loss rate, and AWGN are also able to affect the performance of the system. Finally, the correctness of the results is verified by specific examples.     

大功率LED灯阵能耗测量系统构建

对大功率LED灯阵能耗测量系统进行研究,能够有效解决传统能耗测量系统存在的测量精度低、运行能耗高等问题。依据灯阵能耗测量需求,对大功率LED灯阵能耗测量系统的总体框架进行设计,改进照明管理、通信、随机存储SRAM、Flash闪存、TFT-LCD液晶显示、电源电路和驱动电路等硬件模块,优化开发软件部分的能耗测量功能,完成大功率LED灯阵能耗测量系统的构建。实验结果表明,该系统具有测量精度高、运行能耗低等优点,满足灯阵能耗测量需求。     

A generalized method for the inherent energy performance modeling of machine tools

Machine tools (MTs), as the key equipment of manufacturing systems, have enormous quantities and consume a great amount of energy. However, the diversity of both machines and their energy consumption properties make it difficult to transfer the energy-saving knowledge and services among different MT. To facilitate the initialization configuration of energy-saving services, the inherent energy performance (IEP) is investigated to describe the differences in energy consumption among MTs, and a generalized method for modeling the IEP of MT and its electrical subsystems is proposed. Three key enablers, including generalized experimental design rules, automatic coding, and data processing algorithms, are presented and integrated into a supporting system to reduce the modeling efforts and knowledge requirements. Case studies of an offline manufacturing scenario and an Internet of Things (IoT)-enabled manufacturing scenario were carried out to verify the effectiveness and convenience of the proposed method. The results show that the proposed method can provide essential modeling support for large-scale energy-saving service configurations and energy-efficient MT development.     

Conformational and energy evaluations of novel peptides binding to dengue virus envelope protein

Effective novel peptide inhibitors which targeted the domain III of the dengue envelope (E) protein by blocking dengue virus (DENV) entry into target cells, were identified. The binding affinities of these peptides towards E-protein were evaluated by using a combination of docking and explicit solvent molecular dynamics (MD) simulation methods. The interactions of these complexes were further investigated by using the Molecular Mechanics-Poisson Boltzmann Surface Area (MMPBSA) and Molecular Mechanics Generalized Born Surface Area (MMGBSA) methods. Free energy calculations of the peptides interacting with the E-protein demonstrated that van der Waals (vdW) and electrostatic interactions were the main driving forces stabilizing the complexes. Interestingly, calculated binding free energies showed good agreement with the experimental dissociation constant (K_d) values. Our results also demonstrated that specific residues might play a crucial role in the effective binding interactions. Thus, this study has demonstrated that a combination of docking and molecular dynamics simulations can accelerate the identification process of peptides as potential inhibitors of dengue virus entry into host cells.     

能耗受限时实现性能最大化的作业管理模型

在诸如星载并行计算机这样的系统中,一定周期内,系统能耗存在上限。围绕高产出率计算的概念,针对这种能耗受限条件下性能最大化的作业管理需求,提出了一种作业分配的模型。该模型基于计算节点、作业的性能和能耗参数,动态地进行作业分配,从而在满足能耗上限的前提下获得尽可能大的计算性能。通过模拟实验,验证了该算法的有效性。     

晶体铅动力学及其表面性质的势能函数模拟

采用优化的多体展开势能函数研究面心立方（FCC）晶体铅的动力学和表面性质。在（q00），（qq0)及（qqq)高对称方向，计算声子散射频率与实验曲线良好吻合。Pb(110)表面驰豫严重，其最外层原子向内压缩率达18％，表面7层存在压缩－膨胀交替现象，蒙特卡罗模拟表明，FCC铅在425K开始熔化，模拟结果比纳米铅低约25K。     

水声信号高频小波包能量识别系统模型

对水声信号的识别,提出了高频小波能量法,并构造了一个自动识别系统模型。系统对信号进行小波分解后,取高频子空间能量作为特征向量值进行信号的识别。高频小波能量法可采取小波包与小波两种变换,通过对两种方法的比较显示：高频小波包能量法构造的识别系统能够更好地识别低信噪比水声信号。     

Time and energy modeling of high–performance Level-3 BLAS on x86 architectures

We present accurate piece-wise models for the time and energy costs of high performance implementations of both the matrix multiplication (gemm) and the triangular system solve with multiple right-hand sides (trsm) on x86 architectures. Our methodology decouples the costs due to the floating-point arithmetic/data movement occurring in the higher levels of the cache hierarchy from those of packing/data transfers between the main memory and the L2/L3 cache. A careful analytical study of the data transfers, in combination with an architecture-specific calibration of the costs per operation, render then the components to assemble piece-wise models for the accurate estimation of gemm and trsm’s performance on x86 processors.Our experimental results on an Intel Xeon E5-2620 processor confirm the accuracy of this approach, which reports relative errors for different shapes of gemm and trsm that are, respectively, around 1.5% and 4.5% on average for both time and energy.     

An experimental research on single discharge machining of insulating ceramics efficiently with high energy capacitor

Insulating ceramics are applied to modern manufacturing industries for their improved material properties.But they are the difficult-to-machine materials because of their high rigidity,high brittleness and non-electrical conductivity.A new method which employs a high energy capacitor for electric discharge machining of insulating ceramics efficiently is presented in this paper,and the single discharge experiments have been carried out.The process uses the high voltage,large capacitor and high discharge ener...