基于Kubernetes的无服务器计算与微服务集成架构研究

Kubernetes提供资源调度、应用部署与管理、自动修复、服务发现与负载均衡、弹性伸缩等分布式应用管理功能,屏蔽基础设施差异,成为企业多云管理的操作系统。容器是轻量、可移植、自包含的软件打包技术,能使应用程序在不同平台以相同的方式运行。服务网格将服务治理与业务逻辑解耦,服务治理下沉到基础设施,以无侵入方式实现服务轻量化。无服务器计算是事件驱动、无状态、短运行时间、敏捷自动伸缩和低成本。微服务是去中心、独立、自治、跨语言的小型软件实体集合。以容器为云原生应用的运行载体,用服务网格治理服务,将无服务器计算与微服务相结合,发挥各自优势,为企业提供完备、灵活的云原生应用解决方案。基于Kubernetes的无服务器计算与微服务集成架构为应用提供更好的弹性和灵活性,变革应用的设计、开发和管理模式。     

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。     

遥感影像在线处理流程编排与算法调度技术研究

近年来多载荷、多分辨率的遥感影像数量快速增加,各行业对于遥感产品多样化、高时效性的需求日益增长,传统的生产方式已无法满足需求,亟需构建自动化、批量化的生产模式。为此,该文提出一种基于云原生技术架构的遥感大数据在线处理流程编排与算法调度技术,以容器为算法模型封装载体,以Kubernetes作为底层运行平台,构建流程驱动和资源调度双层架构,支持不同遥感算法的快速集成和用户自定义的生产任务全流程编排,搭建了具备可视化流程设计管理、标准化算法扩展、集群计算运行管理功能的遥感数据处理云服务平台,为遥感大数据批量处理与生产提供高效的解决方案。     

“互联网+”背景下多云架构技术及其应用分析

灵活性对企业来说至关重要,多云战略已经成为越来越多企业的选择。如何将不同公有云服务进行深度整合,实现跨云系统的架构设计和跨云资源的管理调度,日益成为企业上云的新课题。文章对多云部署的架构难点进行了分析,对多云网络、多云聚合的技术细节进行了分析,阐述了目前多云技术及实施要点,并展望了多云应用及生态发展的前景。     

基于Kubernetes的融合云原生基础设施方案与关键技术

随着云原生技术在IT域和CT域的广泛应用,云原生基础设施逐渐成为新一代的云计算基础设施。分析了主流云原生基础设施技术体系及其局限性,提出了基于 Kubernetes 的轻量级融合云原生基础设施及其管理方案,探讨了融合云原生基础设施的关键技术和应用场景。     

基于算力网络平台算力资源处理技术浅析

随着算力网络的发展，如何整合多云算力资源，对外统一提供云计算服务成为重要的研究课题。设计算力网络资源管理平台对网络云、移动云、IT云以及边缘云等多云算力资源进行了统一计算处理，未来可以作为算力感知平台承担基础算力的注册、度量、对外发布等功能。本文阐述了算力网络资源管理平台设计方案，用于高效处理海量算力数据，提供即时算力资源信息的技术解析能力，并提出了基于Hudi数据湖的数据处理演进思路和分布式算力资源管理的设想，为其他研究人员提供参考。     

数据中心云资源池节能方案研究

本文通过研究中国移动网络云资源池承载业务的高容灾资源预留、低弹性响应时延特性，对网络云数据中心能耗构成要素进行分析，在实验环境中验证各要素对数据中心能耗及云上业务的影响。根据实验结果，提出了一种网络云资源池全景式节能方案：通过优化虚拟层资源调度实现计算资源适度集中，减少无效资源占用；通过设备综合节能措施实现设备级能耗灵活管控；通过机房、微模块、设备智能联动温控，降低机房环境能耗。通过上述措施的实施，实现了云数据中心节能方案的闭环，有效降低网络云数据中心用电成本，达到增收减支的效果，同时解决了节电过程中带来的资源冷启动、关联告警等问题，确保上云业务无感知。     

上海市数字政府多云异构管理体系分析与研究

数字政府多云异构管理是国内大型城市政务云运行管理的难点。因此提出了针对大型城市的政务云多云服务体系和政务云多云管理服务技术体系,设计和研究了面向政务云管理的多云服务体系和多云管理服务模型,提出各级云管平台数据从汇集、交换、使用和应用等过程进行技术定制、集成和扩展的管理方案,研究了政务云多云环境下平台提供方、平台使用方、平台服务方和服务监管方一体管控的政务云多云管理服务架构,为国内大型城市的政务云运行管理和服务保障体系建设提供了参考。     

基于列生成的云平台资源约束项目调度研究

随着云计算技术的不断发展,越来越多的实验室以及办公环境都采用云平台来获取计算资源。但是,在对云平台相关技术的研究过程中,发现对于云平台资源约束项目的调度问题一直都是一个比较大的挑战。主要原因就是,对资源约束项目进行调度需要考虑资源利用率以及调度的时间成本。根据问题建立了资源约束项目资源库调度模型和一种基于列生成算法的云平台资源约束项目算法。通过与拉格朗日技术、数字优化技术及自适应遗传算法等进行实验对比。结果表明,该方法在问题的解决上是具有明显优势的,也验证了该方法的有效性。     

面向政府行业多云管理模式下的多形态应用支撑关键技术与应用

随着云计算十余年的发展、容器化技术和微服务架构的广泛应用和普及,基于云计算的应用支撑已经逐渐从服务和资源交付向云原生化价值赋能进行升华,逐步走向以应用为核心的云原生服务应用与支撑模式,以实现行业应用更灵活、更高效、更可靠地支撑和交付。面向政府行业多云管理模式下的多形态应用支撑关键技术,重点解决当前行业异构多云的信息化建设背景下,资源分散、统筹力度不足、利旧困难、云原生应用转型及多形态应用支撑能力薄弱等问题,帮助行业客户实现从传统IT架构向基于云平台的高可靠、高并发、高弹性的云原生架构转型,助力行业应用创新和快速上云。     

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.     

云计算环境下任务调度算法的研究

在云计算环境中存在庞大的任务数,为了能更加高效地完成任务请求,如何进行有效地任务调度是云计算环境下实现按需分配资源的关键。针对调度问题提出了一种基于蚁群优化的任务调度算法,该算法能适应云计算环境下的动态特性,且集成了蚁群算法在处理NP-Hard问题时的优点。该算法旨在减少任务调度完成时间。通过在CloudSim平台进行仿真实验,实验结果表明,改进后的算法能减少任务平均完成时间、并能在云计算环境下有效提高调度效率。     

Scheduling strategy for science workflow with deadline constraint on multi-cloud

In view of the deadline-constrained scientific workflow scheduling on multi-cloud,an adaptive discrete particle swarm optimization with genetic algorithm (ADPSOGA) was proposed,which aimed to minimize the execution cost of workflow while meeting its deadline constrains.Firstly,the data transfer cost,the shutdown and boot time of virtual machines,and the bandwidth fluctuations among different cloud providers were considered by this method.Secondly,in order to avoid the premature convergence of traditional particle swarm optimization (PSO),the randomly two-point crossover operator and randomly one-point mutation operator of the genetic algorithm (GA) was introduced.It could effectively improve the diversity of the population in the process of evolution.Finally,a cost-driven strategy for the deadline-constrained workflow was designed.It both considered the data transfer cost and the computing cost.Experimental results show that the ADPSOGA has better performance in terms of deadline and cost reducing in the fluctuant environment.     

A Multi-objective Optimal Task Scheduling in Cloud Environment Using Cuckoo Particle Swarm Optimization

In cloud computing, varied demands are placed on the constantly changing resources. The task scheduling place very vital role in cloud computing environments, this scheduling process needs to schedule the tasks to virtual machine while reducing the makespan and cost. The task scheduling problem comes under NP hard category. Efficient scheduling method makes cloud computing services better and faster. In general, optimization algorithms are used to solve the scheduling issues in cloud. So, in this paper we combined two optimization algorithms namely called as Cuckoo Search (CS) and Particle Swarm Optimization (PSO).The new proposed hybrid algorithm is called as, CS and particle swarm optimization (CPSO). Our main purpose of the proposed paper is to reduce the makespan, cost and deadline violation rate. The performance of the proposed CPSO algorithm is evaluated using cloudsim toolkit. From the simulation results our proposed works minimize the makespan, cost, deadline violation rate, when compared to PBACO, ACO, MIN–MIN, and FCFS.

     

A cloud database route scheduling method using a hybrid optimization algorithm

Cloud computing has appeared as a technology allowing a company to employ computing resources such as applications, software, and hardware to calculate over the Internet. Scholars have paid great attention to cloud computing because of its cutting-edge availability, cost decrement, and boundless applications. A cloud database is a data storage site on the web where the optimal path is spotted to access the needed database. So, placing the ideal path to a database is crucial. The cloud database defined the scheduling problem to choose the perfect route. Cloud database path scheduling is a multifaceted procedure consisting of congestion control, routing list, and network flow distribution. It has a postponement in searching for the needed source route from the cloud database. Offering numerous infinite resources with the growing database workload is an NP-Hard optimization problem where the query request needs optimal schedules to respond to the required services. So, we have used a hybrid cuckoo search (CS) and genetic algorithm (GA), motivated by a social bird's phenomenon, to solve this problem. Integrating genetic operators has dramatically enhanced the balance between the capability of searching and utilization.     

智能兴趣热点:一种基于室内虚拟云的移动云计算系统研究（英文）

With network developing and virtualization rising,more and more indoor environment(POIs) such as cafe,library,office,even bus and subway can provide plenty of bandwidth and computing resources.Meanwhile many people daily spending much time in them are still suffering from the mobile device with limited resources.This situation implies a novel local cloud computing paradigm in which mobile device can leverage nearby resources to facilitate task execution.In this paper,we implement a mobile local computing system based on indoor virtual cloud.This system mainly contains three key components:1) As to application,we create a parser to generate the "method call and cost tree" and analyze it to identify resourceintensive methods.2) As to mobile device,we design a self-learning execution controller to make offloading decision at runtime.3) As to cloud,we construct a social scheduling based application-isolation virtual cloud model.The evaluation results demonstrate that our system is effective and efficient by evaluating CPUintensive calculation application.Memoryintensive image translation application and 1/O-intensive image downloading application.     

基于云计算平台的智慧城市管理系统设计

为了解决智慧城市管理过程中常出现资源调度速度过慢问题,设计了云计算平台的智慧城市管理系统。该系统采用云管理模块下监控各硬件设备,并构建云计算资源调度目标函数,利用文化粒子群算法对目标函数求解,得到云计算资源调度方案,最后测试结果表明,该系统能够实现智慧城市有效管理,并能实时监测城市情况,在实行资源调度时,任务完成时间较短且系统利用率较高,能够实现资源最大化利用。     

A workflow task scheduling algorithm based on the resources' fuzzy clustering in cloud computing environment

Cloud computing is the key and frontier field of the current domestic and international computer technology, workflow task scheduling plays an important part of cloud computing, which is a policy that maps tasks to appropriate resources to execute. Effective task scheduling is essential for obtaining high performance in cloud environment. In this paper, we present a workflow task scheduling algorithm based on the resources' fuzzy clustering named FCBWTS. The major objective of scheduling is to minimize makespan of the precedence constrained applications, which can be modeled as a directed acyclic graph. In FCBWTS, the resource characteristics of cloud computing are considered, a group of characteristics, which describe the synthetic performance of processing units in the resource system, are defined in this paper. With these characteristics and the execution time influence of the ready task in the critical path, processing unit network is pretreated by fuzzy clustering method in order to realize the reasonable partition of processor network. Therefore, it largely reduces the cost in deciding which processor to execute the current task. Comparison on performance evaluation using both the case data in the recent literature and randomly generated directed acyclic graphs shows that this algorithm has outperformed the HEFT, DLS algorithms both in makespan and scheduling time consumed. Copyright © 2014 John Wiley & Sons, Ltd.     

基于云、网、边融合的边缘计算新方案:算力网络

边缘计算已经成为5G时代重要的创新型业务模式,尤其是其低时延特性,被认为是传统方案所不具备的,因此边缘计算能够提供更多的服务能力且具有更为广泛的应用场景。但边缘计算与处于中心位置的云计算之间的算力协同成为新的技术难题,即需要在边缘计算、云计算以及网络之间实现云网协同、云边协同,甚至边边协同,才能实现资源利用的最优化。在研究边缘计算算力分配和调度需求的基础上,提出了基于云、网、边深度融合的算力网络方案,并针对AI类应用给出了一个典型实施系统,该方案能够有效应对未来业务对计算、存储、网络甚至算法资源的多级部署以及在各级节点之间的灵活调度。