检索结果-南通市图书馆

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

基于ICN的数据中心虚拟机动态迁移方法

电子设计工程 2024年第21期32卷 185-191页

作者：王婷尤佳莉韩锐中国科学院声学研究所国家网络新媒体工程技术研究中心北京100190 中国科学院大学北京100049

为了解决传统数据中心网络在支持虚拟机(Virtual Machine,VM)快速动态迁移过程中不能完全避免服务中断的问题,提出了一种基于信息中心网络技术(Information Centric Networking,ICN)的VM动态迁移方法——ICVM。ICVM基于ICN中标识与地址... 详细信息

为了解决传统数据中心网络在支持虚拟机(Virtual Machine,VM)快速动态迁移过程中不能完全避免服务中断的问题,提出了一种基于信息中心网络技术(Information Centric Networking,ICN)的VM动态迁移方法——ICVM。ICVM基于ICN中标识与地址分离的思想,将VM从地址上解耦,路由请求基于VM标识,因此,当VM被迁移到不同的物理机时,其IP、MAC地址可保持不变,服务不会中断。实验结果表明,相较于目前数据中心网络广泛采用的EVPN-VXLAN技术,ICVM不仅支持VM的无缝动态迁移,而且所产生的控制开销数量平均减少了约98%,网络中交换机转发表大小平均减少了约97%,具备很强的扩展性。

关键词：数据中心网络信息中心网络虚拟机迁移服务连续性

Peer priority based data scheduling algorithm in P2P streaming system

在线全文

学校读者我要写书评

暂无评论

High Technology Letters 2013年第2期19卷 208-213页

作者：任浩 Wang Jinlin you jiali National Network New Media Engineering Research Center Institute of AcousticsChinese Academy of Sciences Graduate University of Chinese Academy of Sciences State Nuclear Power Software Development Center

In order to solve the problem that the existing data scheduling algorithm cannot make full use of neighbors' bandwidth resources when allocating data request among several senders in the multisender based P2P streaming system,a peer priority based scheduling algorithm is *** algorithm calculates neighbors' priority based on peers' historical service evaluation as well as how many wanted data that the neighbor *** data request allocated to each neighbor is adjusted dynamically according to the priority when *** with high priority are preferred to allocate more data *** shows the algorithm can make full use of neighbors' bandwidth resources to transmit data to reduce server pressure effectively and improve system scalability.

关键词： peer-to-peer (P2P) streaming data scheduling priority request allocation

Benefiting from providing multi-dimensional view of peers in overlay of P2P streaming system

在线全文

学校读者我要写书评

暂无评论

High Technology Letters 2013年第2期19卷 214-220页

作者：苏杭 Wang Jinlin you jiali Graduate University of Chinese Academy of Sciences National Network New Media Engineering Research Center Institute of AcousticsChinese Academy of Sciences

Most overlay of existing P2P streaming systems just focus on the view point of video content *** multi-dimensional overlay for the P2P streaming system(MDOPS) is proposed for providing multi-dimensional view including video data,peers' service capability and online stability based on locality sensitive *** organizes all Live/VoD peers and the above multi-dimensional information in a one-dimensinal DHT,uses range resource information publish/search and introduces multiple load balancing *** maintains an additional candidate coordinating peer list with high qualified peers who own the video data the peer would possibly access currently and in *** list could speed up the process of searching peers for data scheduling *** experiment based on trace of real streaming system has testified that MDOPS can effectively improve the quality of search results and smooth load distribution among peers without increasing the cost of resource publish/search.

关键词： Peer-to-Peer (P'2P) streaming overlay locality sensitive hashing rangesearch load balance

MapReduce based computation of the diffusion method in recommender systems

在线全文

学校读者我要写书评

暂无评论

High Technology Letters 2016年第3期22卷 288-296页

作者：彭飞 you jiali Zeng Xuewen Deng Haojiang National Network New Media Engineering Research Center Institute of Acoustics Chinese Academy of Sciences Beijing 100190 P. R China University of Chinese Academy of Sciences Beijing 100049 P. R. China

The performance of existing diffusion-based algorithms in recommender systems is still limited by the processing ability of a single computer. In order to conduct the diffusion computation on large data sets,a parallel implementation of the classic diffusion method on the MapReduce framework is proposed. At first,the diffusion computation is transformed from a summation format to a cascade matrix multiplication format,and then,a parallel matrix multiplication algorithm based on dynamic vector is proposed to reduce the CPU and I / O cost on the MapReduce framework,which can also be applied to other parallel matrix multiplication scenarios. Then,block partitioning is used to further improve the performance,while the order of matrix multiplication is also taken into *** on different kinds of data sets have verified the efficiency of the proposed method.

关键词： MapReduce recommender system diffusion parallel matrix multiplication

面向ICN组播的基于感染力的汇聚点部署方法

在线全文

学校读者我要写书评

暂无评论

网络新媒体技术 2024年第1期13卷 21-28页

作者：田震宇尤佳莉朱小勇中国科学院声学研究所国家网络新媒体工程技术研究中心北京100190 中国科学院大学北京100049

在组播系统中,汇聚点(RP)作为组播树的根节点,负责将源内容引入组播树。RP汇集了网络中的流量,容易成为网络性能瓶颈。为了提高网络容纳的组播组数量(组播容量),可以将更多的网络节点升级为RP。相比于传统的IP组播,信息中心网络(ICN)组... 详细信息

在组播系统中,汇聚点(RP)作为组播树的根节点,负责将源内容引入组播树。RP汇集了网络中的流量,容易成为网络性能瓶颈。为了提高网络容纳的组播组数量(组播容量),可以将更多的网络节点升级为RP。相比于传统的IP组播,信息中心网络(ICN)组播大幅降低了RP管理代价,从而允许部署大量RP。因此,本文基于ICN组播展开研究。但是,由于设备升级成本和系统稳定性等原因,ICN组播中RP数量仍然不能太多。现有基于P中值模型法的RP部署方法没有充分考虑到RP之间的负载压力,容易导致RP计算资源浪费和一些RP的负载过重,限制了组播容量。因此,本文提出了一种面向ICN组播的基于感染力的RP部署方法。该方法首先提出了一种组播感染力指标,用于刻画RP之间的负载压力。其次,利用组播感染力,提出了一种基于贪心策略的RP部署方法,优先将负载压力小的节点升级为RP。2组实验证明,在部署15个RP的情况下,本方法相比P中值模型法,将组播容量提高3.48倍;在组播容量都达到9000组时,本方法所需的RP数量仅为p中值模型法的29.4%。

关键词：组播汇聚点部署组播容量感染力

在线全文

学校读者我要写书评

暂无评论

ICN网内解析自适应地址选择方法

网络新媒体技术 2024年第4期13卷 8-15页

作者：马赵琳尤佳莉邓浩江中国科学院声学研究所国家网络新媒体工程技术研究中心北京100190 中国科学院大学北京100049

在信息中心网络中,名字解析系统负责标识与网络地址之间的映射关系的建立、维护和发布。信息中心网络中内容的多副本特性导致标识到地址的“一对多”的映射关系,带来选择适当解析地址的挑战。网内解析作为一种系统驱动的解析模式,将解... 详细信息

在信息中心网络中,名字解析系统负责标识与网络地址之间的映射关系的建立、维护和发布。信息中心网络中内容的多副本特性导致标识到地址的“一对多”的映射关系,带来选择适当解析地址的挑战。网内解析作为一种系统驱动的解析模式,将解析能力内生于可编程网络设备中,不可避免地需要进行选址操作。现有的选址方法不能同时提供网络和服务端负载均衡的保障,传统的直接选择、轮询等策略造成解析链路拥塞或服务端负载不均衡。为解决上述问题,本文提出一种自适应的地址选择方法,充分利用软件定义网络的网络感知和处理能力,综合网络状态信息和地址解析频率信息,同时实现服务端与网络链路的负载均衡。仿真结果证明该方法在负载均衡方面的优越性。提出的算法为未来的研究提供新的方向与思路。

关键词：信息中心网络解析系统地址选择网络状态自适应方法

MCloudSim:modeling and simulating distributed clouds

在线全文

学校读者我要写书评

暂无评论

High Technology Letters 2014年第2期20卷 213-221页

作者：郑鹏飞 you jiali Wang Jinlin Zeng Xuewen National Network New Media Engineering Research Center Institute of AcousticsChinese Academy of Sciences University of Chinese Academy of Sciences

Many distributed clouds which try to integrate the advantages of centralized clouds and distributed systems have been studied in recent *** cloud simulators focus on large scale data centers with virtualized servers,and cannot meet the needs of distributed cloud *** paper introduces a generalized and extensible simulation framework,named MCloudSim,which is used for modeling and simulating distributed *** has the following characteristics;(1)the overlay protocol in MCloudSim can be freely replaced without affecting the function of other modules;(2) a Petri net workflow theory based task model is proposed to simulate distributed tasks;(3) a distributed cloud can be easily built by assembling and expanding the basic resource entities provided by ***,simulation results of scenarios with a 3-tier central-controlled distributed cloud and a P2 P based cloud prove that MCloudSim has high efficiency and satisfies performance in supporting large scale experiments and different distributed clouds.

关键词： cloud computing distributed cloud P2P cloud simulator Petri net

Exhumation and Deformation of the Daba Shan Orocline as Determined from Modern River Sands Apatite Fission-Track

在线全文

学校读者我要写书评

暂无评论

Journal of Earth Science 2023年第4期34卷 1140-1149页

作者： jiali you Zhao Yang Xiaohui Shi Chuanbo Shen Yunpeng Dong Bin Cheng State Key Laboratory of Continental Dynamics Northwest UniversityXi'an 710069China Key Laboratory of Tectonics and Petroleum Resources China University of GeosciencesWuhan 430074China

The Daba Shan orocline is located at the northeastern margin of the Sichuan Basin and has been inferred as a foreland thrust-fold belt of the Qinling Orogen since the Late Triassic.A complete understanding of rock exhumation history is critical to elucidate how and when this typical orocline structure is *** apatite fission-track dating of modern river sands is employed to reveal the regional exhumation history of the Daba Shan *** age peaks are identified and interpreted as the results of tectonic *** older age peaks at ~150–140 and ~116–86 Ma are agreement with two main shortening deformation episodes of the Yanshanian Movement in the eastern *** other two younger age peaks at ~69 and ~37 Ma support that the Daba Shan was reactivated by the Late Cretaceous to Cenozoic deformation which were likely related to the subduction of the Pacific Ocean and eastward growth of the Tibetan Plateau,*** is worth noting that in contrast to the ~150–140 Ma rapid rock uplift and exhumation,the Middle Cretaceous exhumation(~116–86 Ma) shifted southward and continued to spread to southern tips of the Daba *** exhumation variations in temporal and spatial allow a southward thrust deformation with piggyback style during the Yanshanian.

关键词： detrital thermochronology apatite fission-track dating exhumation history Daba Shan orocline.

在线全文

学校读者我要写书评

暂无评论

层次化解析系统中两级节点协同的副本策略

网络新媒体技术 2024年第3期13卷 8-18页

作者：廉文瀚王劲林尤佳莉中国科学院声学研究所国家网络新媒体工程技术研究中心北京100190 中国科学院大学北京100049 鹏城实验室深圳518055

名字解析系统负责存储维护信息中心网络中名字和地址之间的映射关系,并提供名字解析服务。层次化的解析系统因可扩展性良好而备受关注,但解析请求一旦无法就近完成则需要沿其层次化结构进行长距离转发。当只有少数节点存储热门名字映射... 详细信息

名字解析系统负责存储维护信息中心网络中名字和地址之间的映射关系,并提供名字解析服务。层次化的解析系统因可扩展性良好而备受关注,但解析请求一旦无法就近完成则需要沿其层次化结构进行长距离转发。当只有少数节点存储热门名字映射记录时,长解析时延的问题将更严重。副本策略是提高解析服务性能的重要机制,通常解析节点会向其他节点推送自身存储的映射关系从而扩散副本,但这种策略往往不能精准推送用户最感兴趣的名字,也难以控制副本数目。本文提出一种两级节点协同的副本策略,管理域中的高层级解析节点通过信息收集窗口获知服务范围内用户最感兴趣的名字后,将主动从其他节点请求副本。为底层节点分发副本时,以时延优化的边际收益衡量是否应为每个名字再多放置一个副本,从而自适应地决定不同名字的副本数目和位置。实验结果表明,本文的策略准确捕捉了用户的解析需求偏好,有效地提高了管理域内的解析命中率,解析时延相较于无副本时降低34.1%。

关键词：名字解析系统信息中心网络副本策略解析时延节点协同