Kubernetes在中台化业务系统中的应用与方案

Kubernetes是一种开源的容器管理平台，用于管理多个主机上的容器化应用程序。随着大型企业中台化进程的不断推进，很多业务系统都进行了中台化重构，因此也带来了容器管理问题。本文从实际生产中中台化业务系统的容器管理问题出发，基于对Kubernetes容器集群管理技术的研究，解决了集群中多租户权限管控、多集群集中管理、Kubernetes服务外部网络访问策略和机制、容器集群扩缩容调优、集群时钟同步等方面的容器应用问题，确保了业务系统的高效运行。本方法也适用于类似架构的业务系统，具有一定的推广价值。     

浅谈微服务架构、Docker和Kubernetes

当今社会,云平台已经成为各系统的主流建设方式,微服务、容器(Docker)、Kubernetes这些新技术已经在云平台建设中大规模应用。本文介绍了微服务技术架构,以及目前比较热门的Docker容器化技术和容器化管理平台Kubernetes,并给出了采用Docker+Kubernetes的微服务架构的典型方案。     

基于Kubernetes的企业级容器云平台设计与实践

针对企业数字化转型过程中业务系统研发过程不够灵活、系统上线后运维成本不断提高等问题,提出了基于Kubernetes的企业级容器云平台方案。通过将企业业务系统拆分为不同的微服务模块,结合Kubernetes的容器化部署优势,总结了一套基于容器云平台的业务系统开发、部署以及运维的全生命周期管理方案。经过应用实践证明,该方案在保证平台系统稳定运行的同时,可以有效提高业务系统研发过程的灵活性,提高团队的研发效率,提升系统自动化运维能力,降低系统运维成本。     

Kubernetes容器云平台多租户方案研究与设计

在容器云平台中,租户共享底层的计算、存储、网络等资源,存在租户容器运行和数据安全问题。分析了 Kubernetes 访问控制和资源隔离实现方案基础上,提出了一种基于多租户访问控制模型的容器云平台多租户方案,涵盖多租户管理模型、多租户访问控制、计算资源隔离和网络资源隔离等,可切实提升基于Kubernetes的容器云平台的资源隔离能力,有效降低数据安全隐患。     

基于Kubernetes的企业级容器云平台建设

针对企业信息化系统业务升级、故障日益频繁,运维人员工作负荷不断增加等问题,传统虚拟化平台方案已无法满足现有企业业务建设需求.文章详细介绍Kubernetes技术架构,从Docker技术、容器与虚拟化对比、Kubernetes概念等方面细致剖析Kuber-netes实现原理,设计一种基于Kubernetes+Docker...     

面向容器的集群资源管理系统研究

在云计算日益火爆的今天,云计算领域诞生了一个新兴技术,使得整个计算机行业为之振奋。这就是Docker——一个全新的"容器"。Docker作为轻量级容器能够为云计算3种模型(Saa S,Paa S,Iaa S)之一的Paa S的构建提供可靠的支持。文章以Docker为基础平台,以Google开源的Kubernetes为平台,Harbor为镜像仓库,Etcd为全局索引系统,falcon-agent为监控构建完整的集群资源管理系统,提供大型集群环境部署及管理方案。     

Kubernetes平台下国产GPU的并发容器共享

随着技术的不断发展,机载智能系统正在向智能化发展。相较于虚拟机,轻量灵活且可伸缩的容器技术在安全性和性能均要求较高的机载智能应用中更适用。随着容器技术的不断发展,容器编排系统也迅速崛起,其中Kubernetes已成为主流的容器编排平台。然而目前Kubernetes在GPU管理方面存在不足,多个并发GPU容器无法共享GPU资源。考虑到日益增加的硬件自主可控需求,面向容器技术的国产GPU资源的共享迫在眉睫。针对上述问题,提出了一种名为sharedGPU的通用国产GPU共享方法,所提出的方法面向Kubernetes平台,实现了并发容器之间共享GPU,同时保留容器独占GPU的使用模式。实验结果表明,sharedGPU支持独占和共享两种使用GPU的模式。     

Docker技术在软件开发过程中的应用研究

Docker是一种开源容器级虚拟化技术,基于它可建立PaaS云服务。在软件开发中应用Docker技术能够使程序的部署和运行更加高效。通过研究Docker技术的架构,然后在开发环境中应用Kubernetes自动化容器平台,保证了容器集群的高可用和弹性伸缩。最后基于Docker在开发中的具体工作实现,指出了Docker技术在软件开发、运维中的重要性。     

一种基于OpenStack和Kubernetes的实验云

本文首先简要地介绍了虚拟机技术和容器技术,并对虚拟机和容器技术各自的优缺点进行了对比分析;接着详细介绍了开源云平台OpenStack在虚拟机和容器等技术的探索与实践,阐述了容器编排系统Kubernetes的基础架构,并对OpenStack和Kubernetes各自的优缺点进行了对比分析;然后对不同专业教学实验所需的实操...     

基于Kubernetes的气象私有云自动化部署与优化

容器技术日趋成熟,在应对微服务和系统集约化方面具有巨大优势。基于容器技术,采用谷歌开源的Kubernetes,对私有云平台架构中各个组件进行分析,通过开发一键式部署安装脚本,搭建了基于容器的气象私有云平台。这种自动化部署方式,简化了部署过程、提高了工作效率,降低了手动部署过程中误操作的几率。采用IPVS代替集群默认的Iptables,通过CPU和内存使用效率、新增规则延时、新增规则网络带宽消耗四个方面的对比情况,得出了IPVS作为负载均衡器优化了集群的网络性能。     

基于Quartz的集团化调度任务分布部署研究

集团化应用系统中定时任务需要处理的数据量大、任务数量多,为即时快速对所有数据进行处理,需要将任务进行分布部署运行。通过对Quartz调度框架和集群分布部署技术的分析和研究,提出了集团化应用系统任务调度基于Quartz的集群分布部署方案,通过将Quartz和Spring集成并对Quartz调度任务容器进行扩展,在MemCache上进行服务器状态的更新,解决了大型应用系统定时任务的分布运行、故障处理问题。     

数据库集群技术浅析

在介绍数据库集群定义的基础上,重点阐述了数据库集群的两大体系结构：基于数据库引擎的集群技术和基于数据库网关（中间件）的集群技术,并对数据库集群的核心技术：负载均衡和失效转移进行了分析;最后对数据库集群的两大体系结构进行了比较。     

Where should ICT startup companies be established? Efficiency comparison between cluster types

According to the triple helix model, the government, private companies, and universities play an important role in the development of the regional economy and the growth of clusters. Although balanced interaction with the three players is important in the triple helix model, our study points out that these three players have no choice but to compete. In this study, we conducted meta-frontier analysis (MFA) using three groups, the government, private companies, and universities, which are mentioned in the triple helix model to compare the efficiency of information and communication technology (ICT) startup companies. Therefore, we can see which ICT startup companies have grown up. As a result, the government-led cluster is the most efficient, followed by private company-led cluster and finally university-led cluster. Here, we compared all the clusters of ICT startup companies, unlike previous studies which only focused on a few clusters. Also, we provided another implication that the strengths of startup companies in each cluster can change depending on their geographic location.     

电信运营商对容器技术的探索和思考

容器是直接运行在操作系统内核上,拥有相对隔离、独立资源的运行环境。容器技术的发展符合云计算的本质要求,电信运营商作为云计算领域的深度玩家,必须重视容器技术的发展。简要介绍了容器技术背景,分析了容器技术在电信运营商的潜在应用场景,总结了电信运营商在容器技术方面的初步探索案例,探讨了容器技术在电信运营商转型升级中可能发挥的积极作用。     

A Cluster-Based MAC Protocol for Cognitive Radio Ad Hoc Networks

The cognitive radio (CR) technology is an efficient approach to share the spectrum. But how to access the idle channels effectively is a challenging problem. To solve this problem, we propose a MAC protocol for cluster-based CR ad hoc networks, with the aim of making cluster structure more robust to the primary users’ channel re-occupancy activities, so that the CR users can use the idle spectrum more efficiently. Under this framework, a cluster formation algorithm based on available channels, geographical position and experienced statistics is proposed to maximize the network throughput and maintain the cluster topology stability. The schemes for new CR user joining a cluster and cluster maintenance are also designed. An experience database is proposed to store the experienced statistics which are obtained from spectrum occupancy history. Such a database can be used to support the acts of nodes, such as neighbor discovery and cluster forming. Simulation results reveal that our proposed framework not only outperforms conventional CR MAC protocols in terms of throughput and packet delay, but also generates a smaller number of clusters and has more stable cluster structure with less reclustering time.     

基于密码学的访问控制和加密安全数据库

在数据库安全领域的研究中,"数据库管理者"对数据库安全带来的安全隐患越来越受到研究者的关注,尤其是在基于互联网的网络应用提供商(Internet service provider)模式的数据库系统中,这种安全威胁更为严重.传统的数据库访问控制(存取控制)方法,对此安全隐患不能提供有效的安全防范.密码学的安全数据库技术,因为其基于数学难解问题的计算复杂性,成为解决数据库安全问题的日渐重要的方法.本文分析了迄今为止各种不同的,针对分布式数据库应用的安全威胁,并对密码学安全数据库中的基于密码学的访问控制算法和加密数据库技术进行了综述,并对各类方法的密码学原理,算法特性以及其优缺点进行了分析和描述.     

K-PSO: An improved PSO-based container scheduling algorithm for big data applications

In recent years, Docker container technology is being applied in the field of cloud computing at an explosive speed. The scheduling of Docker container resources has gradually become a research hotspot. Existing big data computing and storage platforms apply with traditional virtual machine technology, which often results in low resource utilization, a long time for flexible scaling and expanding clusters. In this paper, we propose an improved container scheduling algorithm for big data applications named Kubernetes-based particle swarm optimization(K-PSO). Experimental results show that the proposed K-PSO algorithm converges faster than the basic PSO algorithm, and the running time of the algorithm is cut in about half. The K-PSO container scheduling algorithm and algorithm experiment for big data applications are implemented in the Kubernetes container cloud system. Our experimental results show that the node resource utilization rate of the improved scheduling strategy based on K-PSO algorithm is about 20% higher than that of the Kube-scheduler default strategy, balanced QoS priority strategy, ESS strategy, and PSO strategy, while the average I/O performance and average computing performance of Hadoop cluster are not degraded.     

The promise of automatic vehicle identification

Automatic vehicle identification (AVI) technology, which permits fully automatic and unique identification of vehicles at specific interrogation points, offers potential benefits in many fields including fleet control, revenue collection, traffic operations, transportation planning, and safety and law enforcement. Three technologies are currently being used for AVI: optical, microwave, and low power radio frequency. All of the three technologies are currently being used in either operating systems or extensive field testing under operating conditions. The first large scale application of AVI was in the railroad industry to identify automatically rolling stock. The optical AVI techniques used have also been applied in automatic container identification and, in a limited scope application, in the operation of an automatic toll collection system. Microwave AVI techniques have been tested successfully in containerization applications and are currently under test by a major toll road in revenue collection application. Low power radio frequency AVI was tested extensively in the federally-sponsored Automatic Bus Identification Project and is being evaluated for nonstop toll collection application by toll agencies. While the cost effectiveness of AVI will vary to a considerable extent with the technology used and the desired functional requirements, existing AVI technology can provide automatic unique identification of vehicles at a cost which can be justified in many applications. As these applications are being developed, the rapid advances in computer and electronic technology which are inherent in AVI systems, will further reduce initial and operating costs of contemplated AVI systems.     

视频轮播改造在视频会议系统中的应用

为实现视频会议智能化管理的目的,采用了视频轮播改造技术。以两处试点改造项目为例,详细介绍了视频轮播改造技术的框架结构和软件处理流程。视频轮播改造技术应用于视频会议管理中,具有易用性、通用性、创新性等特点。所开发的轮播软件已通过测试并投入现场运行。     

电子政务系统的高可用性应用研究

以采用国产软硬件进行电子政务建设研究及示范应用“863计划”课题为背景，分析了烟台电子政务平台建设中国产服务器操作系统高可用性集成方案：数据库的高可用性、WEB的高可用性以及电子邮件的高可用性，说明了集群技术、心跳技术以及RAID技术在电子政务平台中的实际应用，提出了本系统采用的备份策略和电源管理高可用性的可行方案。