基于线程池的高性能服务器软件的设计和实现

传统的多线程技术,虽然在提高服务器性能发挥了重要作用,但是在实时系统应用中,由于其频繁的创建和销毁线程的系统开销严重影响了系统性能,因此有效降低这部分开销对进一步提高服务器性能尤为重要。文中在分析传统服务器模型的弊端基础上,提出了基于线程池的服务器模型,并分析说明了线程池技术如何提高服务器性能,最后给出了应用C 语言实现该服务器模型的具体方法。     

沉管隧道地震响应的三维非线性数值模拟方法及应用

介绍了沉管隧道的传统抗震分析方法,针对上海某沉管越江隧道的非线性地震响应问题,提出了基于显式有限元和高性能计算上的大规模数值模拟仿真方法。建立了包括地基土、沉管隧道以及柔性接头在内的全三维分析模型,单元数和节点数均超过百万。充分考虑到土壤、橡胶等多个对象的材料非线性,以及土体-沉管间、柔性接头-管段间的接触非线性因素,最后利用动态显示算法及其关键技术进行了基于高性能计算机的大规模地震响应分析。计算结果描述了地基土-沉管隧道体系在地震荷载作用下的相互作用和整体变形,同时对沉管及柔性接头等关键部位进行了深入分析,从而为沉管隧道的抗震设计研究提供参考。     

基于Myrinet/GM的多通道通信

通信子系统对并行系统的计算效率有重要影响,大规模应用对并行平台的通信性能和可用性提出了挑战性的要求.多通道通信技术通过并行采用多路网络链路互连来提高并行系统通信性能和可用性.首先分析了多进程复用网络对通信性能的影响,然后以Myrinet/GM网络平台为基础,提出了基于网络接口层的通信链路动态选择与分配策略,设计和实现了支持多路Myrinet网络并行通信的协议层MNC.MNC支持通信进程平等,充分地利用多路Myrinet网络链路资源.在使用2路Myrinet互连的PC机群平台上,MNC进程间通信带宽相对于单链路提高了约34%,有效地提高了应用层通信性能.     

Grid-Controlled Lightpaths for High Performance Grid Applications

Grid applications call for high performance networking support. One attractive solution is to deploy Grids over optical networks. However, resource management in optical domains is traditionally very rigid and cannot successfully meet the requirements of Grid applications, such as flexible provisioning and configuration. In this paper, we present a customizable resource management solution for optical networks where users can create lightpaths on demand and manage their own network resources. Thanks to a Grid-centric system architecture, lightpath resources can be shared among users and easily integrated with data and computation Grids.     

高性能路由器PIM多播路由系统的设计与实现

“基于网络处理器的分布式硬件转发”的第五代路由器将在新一代互联网尤其是其骨干网络中得到广泛部署，网络流媒体等新型应用的快速发展对网络设备的多播支持能力提 出迫切的要求。本文对高性能路由器多播系统结构，控制板PIM协议软件的主要功能模块、重要数据结构、多播报文转发算法，网络处理器的多播转发表结构、多播转发过程
，控制板对网络处理器多播转发的配置方法和二者的交互行为以及相关的主要技术问题进行了系统的讨论。     

Look Left, Look Right, Look Left Again: An Application of Fractal Symbolic Analysis to Linear Algebra Code Restructuring

Fractal symbolic analysis is a symbolic analysis technique for verifying the legality of program transformations. It is strictly more powerful than dependence analysis; for example, it can be used to verify the legality of blocking LU factorization with pivoting, a task for which dependence analysis is inadequate. In this paper, we show how fractal symbolic analysis can be used to convert between left- and right-looking versions of three kernels of central importance in computational science: triangular solve, Cholesky factorization, and LU factorization with pivoting.     

Quantification of six bioactive compounds in Zhenqi Fuzheng preparation by high-performance liquid chromatography coupled with diode array detector and evaporative light scattering detector

A simple and accurate high-performance liquid chromatography (HPLC) coupled with diode array detector (DAD) and evaporative light scattering detector (ELSD) was established for the determination of six...     

R2D2: A scalable deep learning toolkit for medical imaging segmentation

Deep learning has gained a significant popularity in recent years thanks to its tremendous success across a wide range of relevant fields of applications, including medical image analysis domain in particular. Although convolutional neural networks (CNNs) based medical applications have been providing powerful solutions and revolutionizing medicine, efficiently training of CNNs models is a tedious and challenging task. It is a computationally intensive process taking long time and rare system resources, which represents a significant hindrance to scientific research progress. In order to address this challenge, we propose in this article, R2D2, a scalable intuitive deep learning toolkit for medical imaging semantic segmentation. To the best of our knowledge, the present work is the first that aims to tackle this issue by offering a novel distributed versions of two well-known and widely used CNN segmentation architectures [ie, fully convolutional network (FCN) and U-Net]. We introduce the design and the core building blocks of R2D2. We further present and analyze its experimental evaluation results on two different concrete medical imaging segmentation use cases. R2D2 achieves up to 17.5× and 10.4× speedup than single-node based training of U-Net and FCN, respectively, with a negligible, though still unexpected segmentation accuracy loss. R2D2 offers not only an empirical evidence and investigates in-depth the latest published works but also it facilitates and significantly reduces the effort required by researchers to quickly prototype and easily discover cutting-edge CNN configurations and architectures.