HLognGP: A parallel computation model for GPU clusters

Parallel computation model is an abstraction for the performance characteristics of parallel computers, and should evolve with the development of computational infrastructure. The heterogeneous CPU/Graphics Processing Unit (GPU) systems have been and will be important platforms for scientific computing, which introduces an urgent demand for new parallel computation models targeting this kind of supercomputers. In this research, we propose a parallel computation model called HLog_nGP to abstract the computation and communication features of heterogeneous platforms like TH‐1A. All the substantial parameters of HLog_nGP are in vector form and deal with the new features in GPU clusters. A simplified version HLog₃GP of the proposed model is mapped to a specific GPU cluster and verified with two typical benchmarks. Experimental results show that HLog₃GP outperforms the other two evaluated models and can well model the new particularities of GPU clusters. Copyright © 2015 John Wiley & Sons, Ltd.     

Organic electroluminescent characteristics of PS:NPB composite hole transporting layer

Green organic light-emitting devices with a structure of indium-tin-oxide (ITO)/polystyrene (PS):N, N' -bis-(3-naphthyl)-N, N' -biphenyl-( 1,1' -biphenyl)-4,4' -diamine (NPB)/tris-(8-hydroxyquinoline)-aluminum (Alq3)/Mg:Ag were fabricated. A doping system consisting of small-molecular hole transporting material NPB and polymeric matrix PS was applied as a composite hole transporting layer (HTL), and the thin film preparation was simplified via spin-coating technique. By adjusting the component ratio of the doping system, several devices with different concentration proportion of PS:NPB are constructed. The electroluminescent characteristics of the devices were investigated and discussed. This study demonstrated that the difference of doping concentration of NPB has a remarkable impact on the optoelectronic performance of both HTL and the devices. Optimum device performance can be obtained by choosing a suitable concentration proportion of PS:NPB at 1:1. This study contributes to the construction of composite functional layers of organic light-emitting diode (OLED) devices and to the technical modification.     

空穴传输层NPB对OLED性能的影响

在有机电致发光器件制备工艺中,可以通过控制蒸发速率获得均匀致密的有机膜,在其中引入空穴传输层NPB能改善阳极与有机层的接触,提高载流子的注入和平衡水平,使激子复合率得到有效改善,从而大幅提高器件的发光特性.     

基于“魔方”的不同编程模型下科学应用性能比较

本文基于魔方(曙光5000A)"超并行"系统架构比较了不同并行编程模型下NAS Parallel Benchmark Multi-Zone的性能。结果表明,在单节点共享内存存储架构下OpenMP编程模型能获得更好性能,跨节点采用混合编程模型能获得更好性能,测试结果为百万亿次系统的使用及应用性能的提升起到指导作用。     

NPB厚度对堆叠式白光OLED性能的影响

主要研究了NPB厚度对堆叠式白色有机电致发光器件性能的影响。实验制备了四组结构为ITO/2-TNATA(15 nm)/NPB(Tnm)/ADN(30 nm):TBPe(2%):DCJTB(1%)/Alq3(20 nm)/LiF(1 nm)/Al(100 nm)(其中T分别为15,30,35和40 nm)的OLED器件,比较了不同厚度情况下OLED器件的电致发光特性,结果表明:改变NPB(4,4-N,N-bis-N-1-naphthy1-N-pheny1-amino-bipheny1)的厚度能够明显提高器件的发光亮度和发光效率,并调节载流子复合区域的位置,有效提高载流子的注入效果。同时发光器件的颜色也可通过调节NPB层的厚度来改变,这种器件使用NPB作为空穴传输层显示出了色纯度高、亮度好、效率较高的白光发射,其具有CIE色坐标(x=0.301 6,y=0.338 5),最高亮度和最大发光效率分别达到14 020 cd/m2与2.94 lm/W。     

The current-voltage-temperature characteristics of Al/NPB/p-Si contact

The current-voltage (I-V) characteristics of the Al/NPB/p-Si contact shows rectifying behavior with a potential barrier formed at the contact interface. The barrier height and ideality factor values of 0.65 eV and 1.33 are measured at the forward bias of the diode. The barrier height of the Al/NPB/p-Si diode at room temperature is larger that (∼0.58 eV) of conventional Al/p-Si diode. It reveals the NPB organic film control the carrier transport of the diode at the contact interface. The temperature effect on the I-V measurement is also performed to reveal the junction characteristics. The ideality factor of the Al/NPB/p-Si contact increases with decreasing temperature. And the barrier height decreases with decreasing temperature. The effects are due to the existence of the interface states and the inhomogeneous of the barrier at the junction.     

基于高性能计算的开源云平台性能评估

云计算是一种提供各种IT服务的互联网资源利用的新模式,已经广泛地应用在包括高性能计算的各种领域。然而,虚拟化带来了一些性能开销;同时,不同的云平台实施虚拟化技术的不同,使得在这些云平台上应用高性能计算服务的性能也千差万别。通过HPC Challenge (HPCC) Benchmark和NAS Parallel Benchmark(NPB)分别对CPU、内存、网络、扩展性和高性能计算真实负载进行评估,比较并分析了诸如Nimbus、OpenNebula和OpenStack实施高性能计算的性能,实验显示OpenStack对计算密集型的高性能应用负载表现出较好的性能,因此,OpenStack是实施高性能计算的开源云平台的一个好的选择。     

OpenMP on Networks of Workstations for Software DSMs

This paper describes the implementation of a sizable subset of OpenMP on networks of workstations(NOWs) and the source-to-source OpenMP complier(AutoPar） is used for the JIAJIA home-based shared virtual memory system (SVM).The paper suggests some simple modifications and extensions to the OpenMP standard for the difference between SVM and SMP(symmetric multi processor),at which the OpenMP specification is aimed.The OpenMP translator is based on an automatic paralleization compiler,so it is possible to check the correctness of the semantics of OpenMP programs which is not required in an OpenMP-compliant implementation AutoPar is measured for five applications including both programs from NAS Parallel Benchmarks and real applications on a cluster of eight Pentium Ⅱ PCs connected by a 100 Mbps switched Eternet.The evaluation shows that the parallelization by annotaing OpenMPdirectives is simple and the performance of generatd JIAJIA code is still acceptable on NOWs.     

NPB/Alq3双层有机电致发光器件薄膜厚度与器件性能的优化

研究了以NPB为空穴传输层、Alq₃为发光层的双层异质结有机电致发光器件的薄膜厚度对器件性能的影响.制备了一系列具有不同NPB和Alq₃厚度的器件并测试了其电致发光特性.结果表明,器件电流随Alq₃与NPB厚度变化的关系并不相同.不同有机层厚度双层器件的电流机制符合陷阱电荷限制(TCL)理论,随外加电压的增大,器件电流经历了欧姆电导区、TCL电流区、陷阱电荷限制-空间电荷限制(TCL-SCL)过渡区三个区域的变化.当有机层厚度匹配为NPB(20nm)/Alq₃(50nm)时可以获得性能优良的器件.器件的流明效率-电压关系曲线的变化规律是在低电压区较快达到最大值,然后随电压的增加逐渐降低.     

高性能水利科学计算平台测试分析

高性能计算是解决大规模复杂水利水电系统分析计算的有效途径。高性能计算机是高性能计算的平台和基础,对高性能计算机的计算能力进行测试分析是了解设备性能,提高运算效率,指导专业应用的基础性工作。本文结合中国水科院高性能科学计算平台的升级,利用国际惯用的基准测试算法（LinPACK测试、LLCBench测试和NPB测试）对平台的串行和并行计算能力、并行计算效率等计算指标进行测试分析。分析结果显示了该平台高效的并行计算性能,以及对专业程序可移植性具有良好的支持。