检索结果-南通市图书馆

MilkyWay-2 supercomputer： system and application

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2014年第3期8卷 345-356页

作者： Xiangke LIAO liquan xiao Canqun YANG Yutong LU Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha 410073 China College of Computer National University of Defense Technology Changsha 410073 China

On June 17, 2013, MilkyWay-2 （Tianhe-2） supercomputer was crowned as the fastest supercomputer in the world on the 41th TOP500 list. This paper provides an overview of the MilkyWay-2 project and describes the design of hardware and software systems. The key architecture features of MilkyWay-2 are highlighted, including neo-heterogeneous compute nodes integrating commodity- off-the-shelf processors and accelerators that share similar instruction set architecture, powerful networks that employ proprietary interconnection chips to support the massively parallel message-passing communications, proprietary 16- core processor designed for scientific computing, efficient software stacks that provide high performance file system, emerging programming model for heterogeneous systems, and intelligent system administration. We perform extensive evaluation with wide-ranging applications from LINPACK and Graph500 benchmarks to massively parallel software deployed in the system.

关键词： MilkyWay-2 supercomputer petaflops computing neo-heterogeneous architecture interconnect network heterogeneous programing model system management benchmark optimization performance evaluation

Pinpointing and scheduling access conflicts to improve internal resource utilization in solid-state drives

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2019年第1期13卷 35-50页

作者： Xuchao XIE liquan xiao Dengping WEI Qiong LI Zhenlong SONG Xiongzi GE College of Computer National University of Defense TechnologyChangsha 410073China Department of Computer Science University of MinnesotaMinneapolisMN 55455USA

Modern solid-state drives (SSDs)are integrating more internal resources to achieve higher *** accesses across internal resources can potentially enhance the performance of ***,exploiting parallelism inside SSDs is challenging owing to real-time access *** this paper,we propose a highly parallelizable I/O scheduler (PIOS)to improve internal resource utilization in SSDs from the perspective of I/O ***, we first pinpoint the conflicting flash requests with precision during the address translation in the Flash Translation Layer (FTL).Then,we introduce conflict eliminated requests (CERs)to reorganize the I/O requests in the device-level queue by dispatching conflicting flash requests to different *** to the significant performance discrepancy between flash read and write operations,PIOS employs differentiated scheduling schemes for read and write CER queues to always allocate internal resources to the conflicting CERs that are more *** small dominant size prioritized scheduling policy for the write queue significantly decreases the average write *** high parallelism density prioritized scheduling policy for the read queue better utilizes resources by exploiting internal parallelism *** evaluation results show that the paralle/izable I/O scheduler (PIOS)can accomplish better SSD performance than existing I/O schedulers implemented in both SSD devices and operating systems.

关键词： solid-state drive access conflict I/O scheduler internal resource utilization PIOS

An Enhanced Energy Efficiency Tree-Based Hierarchical Routing Based on Additional Connections

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

International Journal of Communications, Network and System Sciences 2017年第5期10卷 208-218页

作者： Shuangyin Ren Chao Peng Lei He liquan xiao Computer School National University of Defense Technology Changsha China China Areodynamics Research and Develop Center Mianyang China

Instrumented is defined as a key characteristic in the upcoming smart cites, which demands a higher Wireless Sensor Networks (WSNs) performance. Tree-based Hierarchical Routing (THR) is a routing protocol for WSNs static scenarios. THR hierarchically maps WSN area into a routing tree. However, fixed topology and hierarchical architecture leads to unbalanced energy overhead distribution and inflexible routes selection. To uniformly distribute routing energy overhead, reduce routing hops and extend network life cycle, we designed an enhanced THR routing protocol based on additional connections such as Neighbors, Brothers and Nephews. Based on these connections, three actions were introduced which are Shortcut, Overhead Balancing and Adoption. A simulation for different network loads on OMNet++4.1 is issued, and realized this THR routing in Chipcon SmartRF04?EB embedded boards based on Zigbee in a point-to-point topology. According to the simulation results and experiment data, this proposal obviously improved the latency performance, reduced energy overhead and extended the network’s life cycle.

关键词： Wireless Sensor Networks (WSNs) OMNet++ Zigbee Tree Based Hierarchical Routing (THR)

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

对中国英语教学宏观模式的思考

外语教学 2005年第5期26卷 35-42页

作者：肖礼全北京师范大学外文学院北京100875

本文在对上个世纪下叶以来中外四个教学模式评述的基础上,构建了一个新的中国英语教学宏观模式,并借此向中国英语教学改革提出五点建议:树立教师和学生平等教学主体观;实施实体英语教学和虚拟英语教学双轨制;区分英语教学的形式和内容;... 详细信息

本文在对上个世纪下叶以来中外四个教学模式评述的基础上,构建了一个新的中国英语教学宏观模式,并借此向中国英语教学改革提出五点建议:树立教师和学生平等教学主体观;实施实体英语教学和虚拟英语教学双轨制;区分英语教学的形式和内容;教学内容从重工具内容转向重信息内容;把教学结果分为可评估近期和不可评估远期两类。

关键词：英语教学教学模式建议

走出中国英语教学的历史误区——外语教学的哲学思考

在线全文

学校读者我要写书评

暂无评论

外语教学 2004年第1期25卷 67-71页

作者：肖礼全北京师范大学外文学院北京100875

本文讨论了中国英语教学五大误区 :教学理论“外”“二”不分 ;盲目追随国外英语教学潮流 ;理论界的语言学情结 ;过分强调传授英语知识 ,过分关注书面语 ;英语教学等同于语言技能训练。笔者分析了误区产生的原由与影响 ,并提出 :中国英... 详细信息

本文讨论了中国英语教学五大误区 :教学理论“外”“二”不分 ;盲目追随国外英语教学潮流 ;理论界的语言学情结 ;过分强调传授英语知识 ,过分关注书面语 ;英语教学等同于语言技能训练。笔者分析了误区产生的原由与影响 ,并提出 :中国英语教学应是外语教学 ;实践重于理论 ,国情高于一切 ,建立基于中国历史与现实的多层次、多元化教学模式 ;淡化语言学情结 ,语言教学基础理论多元化 ;听说为本 ,读写并重 ,以满足新时代的新要求 ;教学内容是英语教学改革的关键 ,语言的本质属性不是交际 ,而是思维。百科知识更能引发语言学习者的思维 ,因此。

关键词：英语教学误区英语教育改革

在线全文

学校读者我要写书评

暂无评论

光互连网络中资源预留的多级光互连仲裁机制

国防科技大学学报 2016年第5期38卷 39-44页

作者：翦杰赖明澈肖立权国防科技大学计算机学院湖南长沙410073

基于资源预留策略提出一种多级光互连仲裁机制,通过将网络分级实现快速、高效的仲裁。多优先级数据缓存队列的传输节点设计,提供了不同类型流量的差异化传输;通过预约式两级仲裁机制,实现网络的完全公平与100%的高吞吐率。设计并对快速... 详细信息

基于资源预留策略提出一种多级光互连仲裁机制,通过将网络分级实现快速、高效的仲裁。多优先级数据缓存队列的传输节点设计,提供了不同类型流量的差异化传输;通过预约式两级仲裁机制,实现网络的完全公平与100%的高吞吐率。设计并对快速仲裁通道进行了合理布局,极大地缩短了仲裁延迟。仿真结果表明:采用基于资源预留的分级仲裁策略,在多种流量模式下所有节点均获得公平的服务。与Feather Weight相比,分级仲裁策略吞吐率提高17%;与2-pass相比,仲裁延迟减少15%,同时,功耗减少5%。

关键词：多级光互连快速仲裁通道资源预留

在线全文

学校读者我要写书评

暂无评论

高性能计算和数据中心融合网络研究综述

国防科技大学学报 2023年第4期45卷 1-10页

作者：陆平静董德尊赖明澈齐星云熊泽宇曹继军肖立权国防科技大学计算机学院湖南长沙410073

随着高性能计算、大数据处理、云计算和人工智能计算呈融合发展趋势,高性能计算网络和数据中心网络的融合网络成为互连网络发展的重要趋势。分析当前融合网络研究现状;针对当前最具代表性的融合网络进行详细阐述,全面展示该领域的最新... 详细信息

随着高性能计算、大数据处理、云计算和人工智能计算呈融合发展趋势,高性能计算网络和数据中心网络的融合网络成为互连网络发展的重要趋势。分析当前融合网络研究现状;针对当前最具代表性的融合网络进行详细阐述,全面展示该领域的最新技术和动态;提出融合网络面临的技术挑战;基于技术挑战对融合网络的发展趋势进行展望,包括融合网络协议栈设计中融合与分化并存、基于在网计算实现融合网络性能加速、面向新兴应用需求优化融合网络性能。

关键词：高性能计算网络数据中心网络融合网络在网计算