云数据中心基于遗传算法的虚拟机迁移模型

提出云数据中心中基于遗传算法的虚拟机迁移模型GA-VMM（genetic algorithm based virtual machine migration）。GA-VMM在虚拟机迁移的时刻考虑的问题维度优于常见的策略,使虚拟机的分配与迁移更加合理与公平。建立了云端能量消耗与在线虚拟机迁移时间消耗数学模型,通过全局遗传算法来优化虚拟机迁移和放置策略。利用某个企业的大数据中心作为云端测试环境,对比测试GA-VMM迁移模型与已有的虚拟机迁移策略的性能。测试结果表明,GA-VMM迁移模型能够更好地减少物理主机的使用数量和虚拟机的迁移次数,SLA（service level agreement violation）违规基本处于稳定状态;GA-VMM可以降低数据中心能耗,性能优于已有的迁移策略。     

Availability analysis of design configurations to compose virtual performance-optimized data center systems in next-generation cloud data centers

Next-generation cloud data centers are based on software-defined data center infrastructures that promote flexibility, automation, optimization, and scalability. The Redfish standard and the Intel Rack Scale Design technology enable software-defined infrastructure and disaggregate bare-metal compute, storage, and networking resources into virtual pools to dynamically compose resources and create virtual performance-optimized data centers (vPODs) tailored to workload-specific demands. This article proposes four chassis design configurations based on Distributed Management Task Force's Redfish industry standard applied to compose vPOD systems, namely, a fully shared design, partially shared homogeneous design, partially shared heterogeneous design, and not shared design; their main difference is based on the used hardware disaggregation level. Furthermore, we propose models that combine reliability block diagram and stochastic Petri net modeling approaches to represent the complexity of the relationship between the pool of disaggregated hardware resources and their power and cooling sources in a vPOD. These four proposed design configurations were analyzed and compared in terms of availability and component's sensitivity indexes by scaling their configurations considering different data center infrastructure. From the obtained results, we can state that, in general, when one increases the hardware disaggregation, availability is improved. However, after a given point, the availability level of the fully shared, partially shared homogeneous, and partially shared heterogeneous configurations remain almost equal, while the not shared configuration is still able to improve its availability.     

面向数据中心的虚拟机部署及优化算法

虚拟机动态配置是解决数据中心能耗低效的有效方法。针对动态配置过程中的虚拟机部署及优化问题展开研究,提出一种新的面向系统能耗的虚拟机部署算法以及基于主动迁移的优化策略。为了降低系统能耗,新算法采用基于服务器利用率的最佳适配降序算法求解虚拟机部署方案;同时为了适应应用负载的动态变化,新算法启动主动迁移策略对部署方案进行优化,即通过启发式算法在当前部署的基础上搜索使系统能耗更低的优化方案,并根据新部署对虚拟机执行主动迁移。考虑到迁移会导致应用服务质量降级和额外能耗,新算法通过在优化策略中设置基于服务器利用率的启动门限,对虚拟机主动迁移频率进行控制。仿真实验表明,所提算法在系统能耗、虚拟机迁移频率、服务器状态切换频率以及服务质量等多项性能指标上均有显著提高     

A cost-effective scheme supporting adaptive service migration in cloud data center

Cloud computing as an emerging technology promises to provide reliable and available services on demand. However, offering services for mobile requirements without dynamic and adaptive migration may hurt the performance of deployed services. In this paper, we propose MAMOC, a cost-effective approach for selecting the server and migrating services to attain enhanced QoS more economically. The goal of MAMOC is to minimize the total operating cost while guaranteeing the constraints of resource demands, storage capacity, access latency and economies, including selling price and reputation grade. First, we devise an objective optimal model with multi-constraints, describing the relationship among operating cost and the above constraints. Second, a normalized method is adopted to calculate the operating cost for each candidate VM. Then we give a detailed presentation on the online algorithm MAMOC, which determines the optimal server. To evaluate the performance of our proposal, we conducted extensive simulations on three typical network topologies and a realistic data center network. Results show that MAMOC is scalable and robust with the larger scales of requests and VMs in cloud environment. Moreover, MAMOC decreases the competitive ratio by identifying the optimal migration paths, while ensuring the constraints of SLA as satisfying as possible.     

软件定义广域网中控制器部署与交换机动态迁移策略

在软件定义广域网（SD-WAN）部署中，由于广域网（WAN）覆盖范围极大这一特性，单控制器部署策略无论在容量、负载还是安全方面都无法满足其需求，多控制器的部署成为必然趋势。而多控制器部署后整体网络的静态配置很难适应动态的网络流变化，从而造成控制器的负载不均衡，整体网络性能降低。针对上述问题，提出一种多控制器部署算法SC-cSNN，以有效减小控制器和交换机之间的传播时延；并提出一种基于时延、控制器容量以及控制器安全等特征的交换机动态迁移算法，以有效解决控制器超负载问题。仿真实验结果表明，SC-cSNN控制器部署算法的平均最大时延优于现有的基于k-means和基于谱聚类的控制器部署算法，交换机动态迁移算法从多特征的角度有效地解决了SD-WAN控制器负载不均衡的问题。     

Integrated maritime fleet deployment and speed optimization: Case study from RoRo shipping

When planning shipping routes, it is common to use a sequential approach where it is first assumed that each ship sails with a given service speed, and then later during the execution of the routes optimize the sailing speeds along the routes. In this paper we propose a new modeling approach for integrating speed optimization in the planning of shipping routes, as well as a rolling horizon heuristic for solving the combined problem. As a case study we consider a real deployment and routing problem in RoRo-shipping. Computational results show that the rolling horizon heuristic yields good solutions to the integrated problem within reasonable time. It is also shown that significantly better solutions are obtained when speed optimization is integrated with the planning of shipping routes.     

基于私有云和物理机的混合型大数据平台设计及实现

大数据分析技术的广泛应用离不开大数据平台的支撑,构建大数据平台已经是很多企业和机构的重要需求。构建大数据平台需要复杂的系统性的技术,特别是需要考虑系统性能和可扩展性两方面需求。随着数据体量不断增大、用户需求不断增多,规划时的数据平台规模很可能不能满足不断变化的需求。因此,设计了一种混合的大数据平台架构:混合使用物理服务器和私有云云主机的大数据平台。这样就兼顾了性能和可扩展性:由于物理服务器性能一般要高于云上的虚拟机,所以构建在物理服务器上的大数据平台,性能一般要好于构建在私有云上大数据平台;从私有云上启动云服务器非常方便、快捷,所以大数据平台的计算和存储结点可以动态弹性地扩容到私有云上,从而保证高峰期的时候大数据平台仍然可以有充足的处理能力。在生产环境实现了这种混合型设计,在生产环境中的测试也表明了这种设计的有效性。     

云数据中心基于贪心算法的虚拟机迁移策略

为了节省云数据中心的能量消耗,提出了几种基于贪心算法的虚拟机（VM）迁移策略。这些策略将虚拟机迁移过程划分为物理主机状态检测、虚拟机选择和虚拟机放置三个步骤,并分别在虚拟机选择和虚拟机放置步骤中采用贪心算法予以优化。提出的三种迁移策略分别为：最小主机使用效率选择且最大主机使用效率放置算法MinMax_Host_Utilization、最大主机能量使用选择且最小主机能量使用放置算法MaxMin_Host_Power_Usage、最小主机计算能力选择且最大主机计算能力放置算法MinMax_Host_MIPS。针对物理主机处理器使用效率、物理主机能量消耗、物理主机处理器计算能力等指标设置最高或者最低的阈值,参考贪心算法的原理,在指标上超过或者低于这些阈值范围的虚拟机都将进行迁移。利用CloudSim作为云数据中心仿真环境的测试结果表明,基于贪心算法的迁移策略与CloudSim中已存在的静态阈值迁移策略和绝对中位差迁移策略比较起来,总体能量消耗少15%,虚拟机迁移次数少60%,平均SLA违规率低5%。     

基于协同数据库的数据迁移模型研究与实现

随着信息技术的发展和企业间协作的加强,企业的各种异构数据资源之间的协同也成为一个研究的热点。在将这些异构数据源进行协同的同时,不可避免地会出现数据的流动或迁移问题。提出在协同数据库的基础上实现数据迁移的一种模型,分析迁移模型的功能组成与实现,给出关于异构数据迁移的几个建议。建立该模型的意义在于减少协同数据库系统中数据库的数量和降低维护的成本。     

Use of distributed data sources in facility location

Facility location decisions are usually determined by cost and coverage related factors although empirical studies show that such factors as infrastructure, labor conditions and competition also play an important role in practice. The objective of this paper is to develop a multi-objective facility location model accounting for a wide range of factors affecting decision-making. The proposed model selects potential facilities from a set of pre-defined alternative locations according to the number of customers, the number of competitors and real-estate cost criteria. However, that requires large amount of both spatial and non-spatial input data, which could be acquired from distributed data sources over the Internet. Therefore, a computational approach for processing input data and representation of modeling results is elaborated. It is capable of accessing and processing data from heterogeneous spatial and non-spatial data sources. Application of the elaborated data gathering approach and facility location model is demonstrated using an example of fast food restaurants location problem.     

基于模型深度强化学习的数据中心主动地板控制

如何消除数据中心的局部热点是困扰数据中心行业的关键问题之一.本文采用主动地板(AVT)来抑制局部机架热点现象,并将数据中心AVT控制问题抽象为马尔可夫决策过程,设计了基于深度强化学习的主动地板最优控制策略.该策略基于模型深度强化学习方法,克服了传统无模型深度强化学习方法采样效率低的缺陷.大量仿真实验结果表明,与经典无模型(PPO)方法相比,所提出的方法可迅速收敛到最优控制策略,并可以有效抑制机架热点现象.     

一种基于XML映射规则的数据迁移方法设计和实现

数据迁移可以看作是实现从源数据库表到目的数据库表的映射。研究了数据迁移的形式化过程,详细讨论了该过程中存在的几类重要映射,提出了基于XML的通用映射规则表示方法。在此基础上,介绍了通过分析数据库表对象创建脚本自动生成XML形式映射规则框架的方法,并讨论了实际的数据迁移过程。     

基于聚类分流算法的分布式蜜罐系统设计

针对现有的网络安全防御系统主动性不足,对未知类型网络数据的判断速度慢、准确性不高的缺陷,设计了一种应用聚类算法对未知类型数据进行聚类分流的分布式蜜罐系统。在聚类过程中,采用一种改进的聚类中心选择算法,对未知类型网络数据进行模糊聚类,将聚类失败的数据分流到蜜罐中进行特征学习,从而尽早地发现新的攻击类型,减轻蜜罐的监控和记录压力,降低蜜罐被攻破的概率,有利于防御时采用更为有效的防御策略。此系统应用在政府某部门的专网中,实验结果验证了在不明显增加系统计算量的情况下,该聚类算法比平均值聚类算法有更高的聚类成功率。     

Design and optimization of VLC based small-world data centers

The visible light communication (VLC) has the potential to provide dense and fast connectivity at low cost. In this paper, we propose a novel VLC enabled Wireless Small-World Data Center (WSWDC). It employs VLC links to achieve a fully wireless data center network (DCN) across racks for the first time. The using of VLC links eliminates hierarchical switches and inter-rack cables, and thus reducing hardware investment, as well as maintenance cost. More precisely, to simplify the configuration and control operations, we propose three DCN design rationales: (1) fully-wireless, all inter-rack links are wireless; (2) easy-deployable, it is not necessary to change the existing infrastructure inside data center; (3) plug-and-play, no extra centralized control operations are required. Previous proposals, however, cannot achieve the three rationales simultaneously. To this end, we first use regular VLC links to interconnect racks as a regular grid DCN and optimize the rack placement to shorten the average path length and the network diameter. To further exploiting the benefits of VLC links, a few random VLC links are carefully introduced to update the wireless grid DCN as a wireless small-world DCN. To avoid the potential interference among VLC links, we deploy VLC transceivers at different heights on the top of each rack. In this way, VLC links would not interfere with others at each height level. Moreover, we design a greedy but efficient routing method for any pair of racks using their identifiers as inputs. Comprehensive evaluation results indicate that our WSWDC exhibits good topological properties and network performance.     

数据中心中TCP连接建立过程的优化方法

针对数据中心中由于SYN包丢失而引起的TCP连接被延迟从而错过任务时间限制的问题,在无需更换现有设备以及无需修改应用和TCP的前提下,提出一种基于加权随机早期检测（WRED）协议的TCP连接初始化的优化方法。该方法解决了连接优化的三个关键问题：如何识别和标记SYN包,如何在交换机上为SYN包预留空间以及需要预留多少空间。与原TCP相比,优化后TCP连接建立的时间极大地减少。实验表明TCP连接初始化优化方法可以解决任务错过规定时间限制的问题。     

面向金融行业的大数据迁移的研究与实现

现有的金融行业的数据管理模式主要依赖于传统关系型数据库,然而传统架构受到拓展能力和存储性能的限制,难以满足大数据时代快速增长的海量数据量处理的需要。针对金融数据规模大、跨地域、跨系统存储、数据多样化等特点,提出了HiETL大数据迁移管理平台,实现了异构关系型数据库业务系统向Hadoop大数据平台的统一迁移,以及海量数据的集中整合、拓展存储、高效分析查询等一站式管理平台,在保证迁移准确的情况下,其速度可达到3?MB/s。     

混部数据中心负载特征及其任务调度优化分析

随着现代互联网数据中心的规模越来越大,数据中心面临着能耗、可靠性、可管理性与可扩展性等方面的挑战。同时,数据中心承载的服务多样,既有在线Web服务,也有离线批处理任务。在线任务要求较低的延迟,而离线任务要求较高的吞吐量。为了提高服务器利用率,降低数据中心能耗,当前数据中心往往将在线任务和离线任务混合部署到同一个计算集群中。在混部场景下,如何同时满足在线和离线任务的不同要求,是目前面临的关键挑战。分析了阿里巴巴于2018年发布的含有4034台服务器的混部计算集群在8天内的日志数据(cluster-trace-v2018),从静态配置信息、动态混部运行状态、离线批处理作业DAG依赖结构等出发,揭示其负载特征,包括任务倾斜与容器部署的相关关系等,根据任务依赖关系与关键路径,提出了相应的任务调度优化策略。     

火力单元部署优化问题的现状与进展

随着信息技术在军事领域中的运用,以网络中心战为理论的信息化战争变得更加具有优势.火力单元部署是网络化战争中的一个重要研究内容,涉及如何充分利用有限的资源来最大化部署火力单元的防御能力.它是一种多约束非线性优化问题.本文对网络化火控系统中火力单元部署优化问题及其研究方法的进展进行了综述.首先介绍了火力单元部署优化问题的基本研究现状.其次从部署空间、约束条件、目标函数3个方面总结了火力单元部署的数学模型,并讨论了求解火力单元部署问题的优化方法,包括编码、约束处理、多目标处理和求解算法的研究现状.最后阐述了火力单元部署优化中的重点以及发展趋势.     

基于BBO算法的WSN覆盖与连接节点部署方案

针对无线传感器网络中目标节点部署能力差的问题,提出基于生物地理学优化(biogeography-based optimization,BBO)算法的节点部署方案,该方案能够在网络中找到满足K-覆盖和M-连通性要求的传感器节点最佳部署位置。提出的基于BBO的算法为目标节点提供了一种有效的编码方案,通过优化构建的加权多目标函数来获得近似最优解,选择最小数量的合适点P,使得所有目标点在满足K-覆盖的同时,覆盖目标的传感器节点也满足M-连接。仿真结果表明,该方案能够在不同的K和M组合下找到合适位置的最优数,而且与其他技术方案相比,该方案的性能具有明显的优势。