基于OpenStack的清华云平台构建与调度方案设计

从一般云计算的体系结构与清华大学的实际需求出发,利用先进的OpenStack平台,采用分层设计的方法设计实现一个可对云资源进行综合管理的清华云平台。分析了该系统的优势和应具备的主要模块功能,重点研究系统中的资源调度关键技术,提出了一种基于任务调度和负载均衡的策略,并通过对调度方案的实验与分析,验证了该调度策略在保证服务性能和执行效率的基础上能够均衡服务器的资源负载,使云平台处于相对稳定的状态。     

基于OpenStack的高吞吐量路由仿真技术研究

融合了轻量级虚拟化、全虚拟化以及SDN技术的OpenStack云平台具有逼真度高、可扩展性强等优势,可有效满足网络仿真的大规模、高逼真、灵活构建的需求。在网络仿真层面,OpenStack的neutron机制支持三层网络高逼真仿真,可为各虚拟设备的无缝通信提供便利,但其虚拟路由器qrouter只能够转发直连网络的数据,难以构建复杂的仿真网络。为此,基于OpenStack平台,提出了一种基于集中式路由引擎与分布式路由部署的路由仿真技术。实验表明,该方法不仅可以基于qrouter构建复杂的仿真网络,而且具有高吞吐量的优势。     

基于OpenStack的私有云平台设计

云计算是当前计算机领域非常热门的研究方向,在此背景下,阐述了云计算、私有云的相关概念及其应用。详细介绍了开源云计算框架Open Stack及其核心组件。提出了一个基于Open Stack的私有云平台架构方案,在实际工作中部署了私有云平台并对相应的计算、存储、网络、镜像等资源进行管理。     

基于OpenStack和Cloudify的自伸缩云平台体系

现有Web服务在面临高并发请求时,会出现响应时间增加,甚至服务器宕机的问题,为此提出一种基于云计算的自伸缩分布式Web系统架构。该架构构建在OpenStack基础设施即服务(IaaS)的平台之上,结合Cloudify平台即服务(PaaS)的平台,实现了自伸缩云应用平台(ECAP);并提出以虚拟机资源模板为标度值来构建模糊矩阵,实现模糊层次分析调度算法。最后在云平台上上传测试应用,并使用压力测试工具对平台进行了测试分析,结果表明所构建平台比普通应用服务器在应用的平均响应时间和负荷性能上表现更好。     

基于GlusterFS的OpenStack平台设计

将GlusterFS应用于OpenStack中替代swift模块,并针对读写性能进行测试,实验证明在GlusterFS下虚拟机读写性能指标IOPS明显提高。     

基于OpenStack的对象存储性能实验及研究

随着云计算的不断发展,基于OpenStack的开源云得到了国内外IT厂商的广泛关注。从服务响应时间和服务吞吐量两个维度来对比万兆网卡和千兆网卡对OpenStack Swift对象存储方案性能的影响。在此基础上,模拟Swift采用万兆网卡适配器后在各种场景下的性能表现。进一步,采用固态硬盘检验其对Swift存储性能的影响。最后进行代理节点和存储节点的配比实验,挖掘云存储技术的价值,设计更加符合最终用户需要的云存储解决方案。     

VM consolidation: A real case based on OpenStack Cloud

In recent years, Cloud computing has been emerging as the next big revolution in both computer networks and Web provisioning. Because of raised expectations, several vendors, such as Amazon and IBM, started designing, developing, and deploying Cloud solutions to optimize the usage of their own data centers, and some open-source solutions are also underway, such as Eucalyptus and OpenStack. Cloud architectures exploit virtualization techniques to provision multiple Virtual Machines (VMs) on the same physical host, so as to efficiently use available resources, for instance, to consolidate VMs in the minimal number of physical servers to reduce the runtime power consumption. VM consolidation has to carefully consider the aggregated resource consumption of co-located VMs, in order to avoid performance reductions and Service Level Agreement (SLA) violations. While various works have already treated the VM consolidation problem from a theoretical perspective, this paper focuses on it from a more practical viewpoint, with specific attention on the consolidation aspects related to power, CPU, and networking resource sharing. Moreover, the paper proposes a Cloud management platform to optimize VM consolidation along three main dimensions, namely power consumption, host resources, and networking. Reported experimental results point out that interferences between co-located VMs have to be carefully considered to avoid placement solutions that, although being feasible from a more theoretical viewpoint, cannot ensure VM provisioning with SLA guarantees.     

Dandelion:OpenStack云平台的快速部署机制

提出了一种快速、自动部署OpenStack云平台的解决方案,以提高OpenStack的部署效率.该方案首先创建不同节点类型的镜像模板文件; 接着根据节点类型(如网络节点、计算节点)将已制作好的镜像模板进行复制,并依照节点的属性(如IP地址、hostname标识等)使用脚本自动修改配置文件,完成单个节点的部署; 之后,利用相同的策略实现其他节点的快速部署.在部署的基础上,通过管理服务器提供的预启动执行环境(PXE)、动态主机配置协议(DHCP)及简单文件传输协议(TFTP)等网络服务挂载节点的镜像块文件,最终完成节点的启动.此外,建立了性能评价模型用于确定源镜像副本和存储服务器的最佳数量,以优化存储网络的拓扑结构.实验结果表明,无论是使用相同存储网络部署不同大小的云平台,还是使用不同大小的存储网络部署相同大小的云平台,与Cobbler、网络文件系统(NFS)等部署方案相比,所提解决方案大大减少了部署时间,提高了部署效率.     

基于OpenStack资源监控系统

资源监控是提高云平台可靠性的重要手段. 本文结合OpenStack云平台的特点,设计并实现了一个全面、智能、高效的资源监控系统,完成了资源监控系统整体架构的设计以及各模块的功能划分,并给出了实例监控方式、数据存储模型等具体的实现方法. 最后,通过测试结果说明资源监控系统的有效性.     

An integration scheme for image segmentation and labeling based onMarkov random field model

This paper presents a unified approach for the image understanding problem based on the Markov random field models. In the proposed scheme, the image segmentation and interpretation processes cooperate in the simultaneous optimization process so that the erroneous segmentation and misinterpretation can be compensatedly recovered by continuous estimation of the unified energy function     

新的高性能CRC查表算法

针对通用循环冗余检验码(CRC)查表算法,提出了一种基于查表法原理的高性能算法,并对通用查表算法进行了优化。该算法主要是利用了共用体和流水线的性质,通过共用体的性质将原先需两次分别读入的数据一次读入,并去掉了繁琐的移位步骤。而通过流水线的性质实现一次读入数据后,分开算出CRC,然后进行合并,达到一次处理的效果。二者均达到了简化步骤的目的。在进行了理论分析和实验验证后,结果表明该算法运行时间短。     

一种基于分片包标记的改进方案

分布式拒绝服务（DDoS）攻击已经对Internet安全构成巨大威胁。由于TCP/IP协议本身的缺陷以及Internet的无状态性,使受害者对攻击源的确定变得十分困难。在深入研究分片包标记方案的基础上,扩展了标记空间,设立了一个分组域来区分数据包来自于哪一分组的路由器。这样,在重构攻击路径时只需要少量的分片组合就可以验证一条边是否在实际的攻击路径中,从而缩短了收敛时间,并减少了误报数。     

基于OpenStack的Swift负载均衡算法

为了解决由于OpenStack的负载分发不均衡而引发的存储性能下降、资源利用率降低、I/O响应时长增加等问题,提出对加权最小连接调度算法进行改进. 通过对对象存储的负载均衡调度算法研究,利用存储节点的CPU、内存、硬盘、I/O资源利用率信息,并结合节点任务请求连接数,计算存储节点负载能力、性能和权值. 负载均衡器根据每个存储节点的权值大小判断任务分发方向. 经实验证明改进的负载均衡调度算法能够解决存储读写性能下降的问题,提升数据吞吐率、存储读写性能和系统稳定性.     

基于Ceph的OpenStack存储克隆链

当OpenStack使用Ceph作为存储后端时,由于Ceph内部本身克隆机制,导致OpenStack云平台存储管理容易出错且效率低。为了解决和优化OpenStack使用Ceph作为存储后端时在存储管理技术方面的不足,本文在开源云平台OpenStack的基础上,通过后端存储Ceph查询各对象在后端的表现形式,递归查找卷或快照的子链路,实现OpenStack存储克隆链功能,并通过Django和Angular JS实现克隆链可视化。通过克隆链展示,可以清晰地了解OpenStack各存储对象之间的关系,提高OpenStack存储管理效率。      

针对OpenStack平台逻辑架构的研究

OpenStack是一个基于IaaS(Infrastructure as a service)层的开源云计算平台。文章在了解云计算技术的基础上针对OpenStack进行了分析研究,使云计算系统搭建者了解OpenStack整体的工作原理与具体逻辑,对云平台的理解与环境搭建也具有重要的意义。     

基于 OpenStack 的虚拟资源调度技术研究

传统的高能物理计算系统基于物理机集群,主要通过 Torque、Condor、LSF 等资源管理和作业调度系统将作业调度到物理机器上运行,缺少对虚拟化计算的接口支持。我们选取 OpenStack 作为底层虚拟化平台,设计并实现上层调度系统与 OpenStack 之间的桥梁,采用推拉结合的作业运行方式,构建虚拟计算集群。     

An extended processing scheme for coherent integration and parameter estimation based on matched filtering in passive radar

In passive radars, coherent integration is an essential method to achieve processing gain for target detection. The cross ambiguity function （CAF） and the method based on matched filtering are the most common approaches. The method based on matched filtering is an approximation to CAF and the procedure is： （1） divide the signal into snapshots; （2） perform matched filtering on each snapshot; （3） perform fast Fourier transform （FFT） across the snapshots. The matched filtering method is com- putationally affordable and can offer savings of an order of 1000 times in execution speed over that of CAF. However, matched filtering suffers from severe energy loss for high speed targets. In this paper we concentrate mainly on the matched filtering method and we use keystone transform to rectify range migration. Several factors affecting the performance of coherent integration are discussed based on the matched filtering method and keystone transform. Modified methods are introduced to improve the performance by analyzing the impacts of mismatching, precision of the keystone transform, and discretization. The modified discrete chirp Fourier transform （MDCFT） is adopted to rectify the Doppler expansion in a multi-target scenario. A novel velocity estimation method is proposed, and an extended processing scheme presented. Simulations show that the proposed algorithms improve the performance of matched filtering for high speed targets.     

Flexible multibody simulation using hybrid integration scheme

One issue in the dynamic simulation of flexible multibody system is poor computation efficiency, which is due to high frequency components in the solution associated with a deformable body. Standard explicit numerical methods should take very small time steps in order to satisfy the absolute stability condition for the high frequency components and, in turn, the computational efficiency deteriorates. In this study, a hybrid integration scheme is applied to solve the equations of motion of a flexible multibody system for achieving better computational efficiency. The computation times and simulation results are compared between the hybrid scheme and conventional methods. The results demonstrate that the efficiency of a flexible multibody simulation can be improved by using the hybrid scheme.