Design, Analysis, and Implementation of a Novel Low Complexity Scheduler for Joint Resource Allocation

Over the past decade, the problem of fair bandwidth allocation among contending traffic flows on a link has been extensively researched. However, as these flows traverse a computer network, they share different kinds of resources (e.g., links, buffers, router CPU). The ultimate goal should hence be overall fairness in the allocation of multiple resources rather than a specific resource. Moreover, conventional resource scheduling algorithms depend strongly upon the assumption of prior knowledge of network parameters and cannot handle variations or lack of information about these parameters. In this paper, we present a novel scheduler called the composite bandwidth and CPU scheduler (CBCS), which jointly allocates the fair share of the link bandwidth as well as processing resource to all competing flows. CBCS also uses a simple and adaptive online prediction scheme for reliably estimating the processing times of the incoming data packets. Analytically, we prove that CBCS is efficient, with a per-packet work complexity of O(1). Finally, we present simulation results and experimental outcomes from a real-world implementation of CBCS on an Intel IXP 2400 network processor. Our results highlight the improved performance achieved by CBCS and demonstrate the ease with which it can be implemented on off-the-shelf hardware     

主动网络中CPU和带宽资源分配的公平性

传统的包调度方案能实现单一资源分配的公平性,但不能直接应用于主动网络中的主动节点上,因为主动节点包含多种相互关联的资源。提出一种能实现在主动节点中公平分配资源的包调度算法。仿真实验结果显示,该算法能在主动流之间实现CPU和带宽资源的有效公平分配。     

基于自组织劳动分工的边云协同任务调度与资源缓存算法

针对边缘计算环境中,边缘设备的计算和存储资源有限的问题,探讨高效的边云协同任务调度和资源缓存策略,研究自组织劳动分工群智能算法模型机理,并以此为基础,提出基于蜂群劳动分工“激发-抑制”模型的边云协同任务调度算法(edge cloud collaborative task scheduling algorithm based on bee colony labor division‘activator-inhibitor’ model, ECCTS-BCLDAI)和基于蚁群劳动分工“刺激-响应”模型的边云协同资源缓存算法(edge cloud collaborative resource caching algorithm based on ant colony labor division ‘stimulus-response’ model,ECCRC-ACLDSR).仿真实验结果表明:所提出的ECCTS-BCLDAI任务调度算法在降低平均任务执行时长、减少边云协同费用上相较于传统算法有更好的表现;所提出的ECCRC-ACLDSR资源缓存算法在降低任务平均时长、优化网络带宽占用率、减少...     

Priority scheduling for networked virtual environments

When simulating large virtual environments (VEs), contention for limited resources such as the CPU, rendering pipeline, or network bandwidth frequently degrades the system's performance. Whenever such a competition occurs and not all elements that require the resource can be serviced, an approximation must be made to avoid compromising interactive performance. We propose an enhancement to the round-robin approach called Priority Round Robin scheduling (PRR). This algorithm enforces priorities, while retaining the output sensitivity and starvation-free performance of round-robin scheduling. Priorities are set by a user-defined error metric (such as visual error), which the algorithm attempts to minimize. This permits not only scheduling the entities competing for a resource such as updates competing for the network, but also filling the gap between filtering and scheduling techniques. We evaluate our algorithm in a client-server system     

面向移动云计算的自适应虚拟机调度算法

随着移动云计算的快速发展和应用普及,如何对移动云中心资源进行有效管理同时又降低能耗、确保资源高可用是目前移动云计算数据中心的热点问题之一.本文从CPU、内存、网络带宽和磁盘四个维度,建立了基于多目标优化的虚拟机调度模型VMSM-EUN(Virtual Machine Scheduling Model based on Energy consumption,Utility and minimum Number of servers),将最小化数据中心能耗、最大化数据中心效用以及最小化服务器数量作为调度目标.设计了基于改进粒子群的自适应参数调整的虚拟机调度算法VMSA-IPSO(Virtual Machine Scheduling Algorithm based on Improved Particle Swarm Optimization)来求解该模型.最后通过仿真实验验证了本文提出的调度算法的可行性与有效性.对比实验结果表明,本文设计的基于改进粒子群的自适应虚拟机调度算法在进行虚拟机调度时,能在降低能耗的同时提高数据中心效用.     

基于动态关键路径的仿真网格资源调度算法

从仿真系统工作流的关键路径分析，确定关键路径上的联邦成员节点，使网格资源调度方面优先获得保证。簦于仿真网格系统运行的复杂性和不确定性，引入了随机规划理论，提出了仿真网格的动态关键路径概念以及基于动态关键路径的资源调度算法，共同解决当资源有限以及任务相关情况下仿真网格的资源调度问题。具体实现，主要使用遗传算法以及MCP算法。仿真实验结果表明，基于动态关键路径的资源调度算法能够优化仿真网格资源的调度。     

时间敏感网络业务流动态调度场景中的批量重配置算法

工业自动化领域广泛使用时间敏感网络技术. 该领域业务流的调度方式主要包含静态调度和动态调度. 静态调度一次计算所有业务流, 可以最大程度节省链路和时间资源, 但是计算时间长, 无法灵活处理新增业务流. 动态调度以增量的形式计算新增业务流, 计算时间短, 但是资源分配不够合理, 会产生时隙碎片. 全局流重配置机制可以定期对网络中所有业务流进行重新规划, 来优化链路和时间资源的分配, 但该机制只适用于拥有较少业务流的小型网络, 业务流数量的增多会引起计算时间的急剧增长, 影响后续到来的业务流. 本文在现有动态调度算法的基础上, 设计了批量重配置算法. 该算法给出了新的评价指标——网络吞吐率, 并在满足动态调度秒级响应时间的情况下, 定期重配置网络中的部分业务流, 优化网络资源配置. 此外, 算法给出了重配置业务流的选取标准, 并优化了流的路径选择标准和传输开始时间计算方式. 本文针对原算法和增加了批量重配置机制的改进算法进行了仿真实验, 实验结果表明, 改进算法可以在拥有数千条业务流的大型网络运行, 并在网络吞吐率和调度成功的流数量方面有16.5%和5.5%的提升, 同时保证了算法的秒级计算时间.     

基于遗传算法的Kubernetes资源调度算法

Kubernetes在优选阶段仅根据节点CPU和内存的利用率来决定节点的分值,这只能保证单节点的资源利用率,无法保证集群资源的负载均衡.针对该问题,提出一种基于遗传算法的Kubernetes资源调度算法,该算法加入了网络带宽和磁盘IO两项评价指标,同时为评价指标赋予不同权重值,并且引入校验字典校验并修复遗传算法生成的新种群中不符合配置的个体.实验结果表明,与Kubernetes默认资源调度策略相比,该算法考虑了集群中的所有节点的资源利用率,在保证集群负载均衡方面有着更好的效果.     

Large matrix–vector products on distributed bus networks with communication delays using the divisible load paradigm: performance analysis and simulation

We present a performance analysis and experimental simulation results on the problem of scheduling a divisible load on a bus network. In general, the computing requirement of a divisible load is CPU intensive and demands multiple processing nodes for efficient processing. We consider the problem of scheduling a very large matrix–vector product computation on a bus network consisting of a homogeneous set of processors. The experiment was conducted on a PC-based networking environment consisting of Pentium II machines arranged in a bus topology. We present a theoretical analysis and verify these findings on the experimental test-bed. We also developed a software support system with flexibility in terms of scalability of the network and the load size. We present a detailed discussion on the experimental results providing directions for possible future extensions of this work.     

基于改进秃鹰搜索算法的Kubernetes资源调度应用

Kubernetes是一个管理容器化应用的开源平台,其默认的调度算法在优选阶段仅把CPU和内存两种资源来作为计算节点的评分指标,同时还忽略了不同类型的Pod对节点资源的占用比例是不同的,容易导致某一资源达到性能瓶颈,从而造成节点对资源使用失衡.针对上述问题,本文在Kubernetes原有的资源指标基础上增加了带宽和磁盘容量,考虑到CPU、内存、带宽和磁盘容量这4类资源在节点上的占用比例对节点的性能的影响,可能造成Pod中应用的非正常运行,甚至杀死Pod,从而影响集群整体的高可靠性.本文将等待创建的Pod区分为可压缩消耗型、不可压缩消耗型以及均衡型,并为每种类型的Pod设置相应的权重,最后通过改进的秃鹰搜索算法(TBESK)来寻找出最优节点进行调度.实验结果表明,随着集群中Pod的数量在不断增加,在集群负载较大的情况下, TBESK算法的综合负载标准差和默认的调度算法相比提升了24%.     

Fair and optimal dynamic admission control of elastic flows

The task of assigning part of the forwarding capability of a router to different flows, usually called admission control, is considered and an algorithm to handle the requests is developed. The idea is to admit not only the acceptance and the rejection answers but also a third kind of answer that occurs when there is no available share of the resource at the instant of the request but there may be a quote of resource available in a determined time-window in the future, provided that the active flows are suitably decreased. The algorithm guarantees both the fair occupancy of the resource and the optimality of its usage. Moreover, the only information on which the algorithm relies is the number of flows, and for each one, the minimum bandwidth needed and the desired bandwidth.     

Openflow Based Dynamic Flow Scheduling with Multipath for Data Center Networks

The routing mechanism in Data Center networks can affect network performance and latency significantly. Hash-based method, such as ECMP (Equal-Cost Multi-Path), has been widely used in Data Center networks to fulfill the requirement of load balance. However, ECMP statically maps one flow to a path by a hash method, which results in some paths overloaded while others remain underutilized. Some dynamic flow scheduling schemes choose the most underutilized link as the next hop to better utilize the network bandwidth, while these schemes lacks of utilizing the global state of the network. To achieve high bandwidth utilization and low latency, we present a dynamic flow scheduling mechanism based on OpenFlow protocol which enables monitoring the global network information by a centralized controller. Depending on the network statistics obtained by the OpenFlow controller, the routing algorithm chooses the best path for the flow. Because there are two kinds of flows in a Data Center, short-lived flows and long-lived flows, we proposed two different algorithms for them. The implementation uses pox as OpenFlow controller and mininet as the network emulator. The evaluation results demonstrate that our dynamic flow scheduling algorithm is effective and can achieve high link utilization.     

一种可重构资源管理模型及其调度技术

针对多任务操作系统的可重构资源管理,提出了一种管理模型和在线调度算法,具体实现了把任务分配给基于块划分的可重构器件。一方面,可重构器件由一个主CPU控制,主CPU运行在线调度器和放置器;另一方面,可重构器件由具有相同垂直尺寸的固定大小的块构成,但块可以有不同的宽度,目的是为了在资源和任务之间实现更好的匹配;同时在在线调度器和放置器运行两个函数fSPLIT和fSELECT来实现任务在可重构器件上的配置和调度。仿真结果表明,提出的资源管理模型和调度算法不仅能够实现任务集平均响应时间的最小化和有效调度,而且相比于其他调度算法,还能获得更高的资源利用率。     

Bandwidth‐aware divisible task scheduling for cloud computing

Task scheduling is a fundamental issue in achieving high efficiency in cloud computing. However, it is a big challenge for efficient scheduling algorithm design and implementation (as general scheduling problem is NP‐complete). Most existing task‐scheduling methods of cloud computing only consider task resource requirements for CPU and memory, without considering bandwidth requirements. In order to obtain better performance, in this paper, we propose a bandwidth‐aware algorithm for divisible task scheduling in cloud‐computing environments. A nonlinear programming model for the divisible task‐scheduling problem under the bounded multi‐port model is presented. By solving this model, the optimized allocation scheme that determines proper number of tasks assigned to each virtual resource node is obtained. On the basis of the optimized allocation scheme, a heuristic algorithm for divisible load scheduling, called bandwidth‐aware task‐scheduling (BATS) algorithm, is proposed. The performance of algorithm is evaluated using CloudSim toolkit. Experimental result shows that, compared with the fair‐based task‐scheduling algorithm, the bandwidth‐only task‐scheduling algorithm, and the computation‐only task‐scheduling algorithm, the proposed algorithm (BATS) has better performance. Copyright © 2012 John Wiley & Sons, Ltd.     

High reliable real-time bandwidth scheduling for virtual machines with hidden Markov predicting in telehealth platform

Reliable and high-performance resource scheduling for Virtual Machines (VMs) in cloud can guarantee the efficiency of remote rescue with telehealth system. When a local disaster, e.g. earthquake and tsunami, happens in a densely populated area, the surging health care demand leads to the increasing workload in Data Centers (DCs) by storing and delivering a mass of patients’ information and real-time physiology signals. However, the current self-adaptive scheduling methods cannot provide a high-accuracy recognizing of the two conditions: urgency or normal, which would procrastinate the system into a high-performance status, while the best rescue time is lost. In this paper, we propose a Primary Node-based architecture for typical telehealth service on cloud, which takes into account both storage and delivery efficiency. We also design a novel algorithm to predicting and allocating the future bandwidth of all VMs in the telehealth service context. This method is able to dynamically adjust each parameter of a Hidden Markov Model (HMM) through collecting the historical information of the bandwidth workload. After we predict the future bandwidth consumption of VMs, a high-performance scheduling method is used to adjust the bandwidth to each VM for health care service. The simulation results prove that this algorithm provides a high-accurate prediction, which guides the allocating module to make decision before the request burst comes. Nevertheless, our algorithm improves the reliability of telehealth services for storing and delivering patients’ information among DCs.     

Kubernetes异构资源细粒度调度策略的设计与实现

在异构资源环境中高效利用计算资源是提升任务效率和集群利用率的关键。Kuberentes作为容器编排领域的首选方案,在异构资源调度场景下调度器缺少GPU细粒度信息无法满足用户自定义需求,并且CPU/GPU节点混合部署下调度器无法感知异构资源从而导致资源竞争。综合考虑异构资源在节点上的分布及其硬件状态,提出一种基于Kubernetes的CPU/GPU异构资源细粒度调度策略。利用设备插件机制收集每个节点上GPU的详细信息,并将GPU资源指标提交给调度算法。在原有CPU和内存过滤算法的基础上,增加自定义GPU信息的过滤,从而筛选出符合用户细粒度需求的节点。针对CPU/GPU节点混合部署的情况,改进调度器的打分算法,动态感知应用类型,对CPU和GPU应用分别采用负载均衡算法和最小最合适算法,保证异构资源调度策略对不同类型应用的正确调度,并且在CPU资源不足的情况下充分利用GPU节点的碎片资源。通过对GPU细粒度调度和CPU/GPU节点混合部署情况下的调度效果进行实验验证,结果表明该策略能够有效进行GPU调度并且避免资源竞争。     

异构云环境下基于分簇的云资源感知任务调度方案

针对提高异构云平台中资源调度的效率,提出了一种基于任务和资源分簇的异构云计算平台任务调度方案。利用K-means算法,根据任务的CPU和I/O处理时间对任务分簇,根据资源的计算能力对资源分簇;然后,将任务簇对应到合适的资源簇,并利用最早截止时间优先（EDF）算法对任务簇中的独立任务进行调度,利用提出的改进型最小关键路径（MCP）算法对依赖性任务进行调度。实验结果表明,在资源异构的云计算环境中,该方案执行任务时间短、能耗低。     

一种基于PID反馈控制的分时调度算法

近年来，在实时操作系统的研究中，已开始将反馈控制融人操作系统的任务调度，将这一想法与分时操作系统相结合，提出一种基于PID反馈控制的分时调度算法，它能根据任务对CPU带宽的个性化需求，并考虑开放的分时系统中CPU占用率的不确定动态变化特性，动态地分配CPU带宽，使CPU对任务集的处理效果始终处于最佳状态，性能分析实验结果表明，它不仅兼容传统的分时调度架构，还具有良好的动态、静态特性及鲁棒性，且引入的调度费用较低。     

Ontology-Based Semantic Priority Scheduling for Multi-domain Active Measurements

Network control and management techniques (e.g., dynamic path switching and on-demand bandwidth provisioning) rely on active measurements of the end-to-end network status. The measurements are needed to meet network monitoring objectives such as network weather forecasting, anomaly detection, and fault-diagnosis. Recent widespread deployment of openly accessible multi-domain active measurement frameworks, such as perfSONAR, has resulted in users competing for system and network measurement resources. Hence, there is a need to prioritize measurement requests of users before they are scheduled on measurement resources. In this paper, we present a novel ontology-based semantic priority scheduling algorithm (SPS) that handles resource contention while servicing measurement requests for meeting network monitoring objectives. We adopt ontologies to formalize semantic definitions and develop an inference engine to dynamically prioritize measurement requests. The prioritization is based upon user roles, user sampling preferences, resource policies, and oversampling mitigation factors. Performance evaluation results demonstrate that our SPS algorithm outperforms existing deterministic and heuristic algorithms in terms of user ‘satisfaction ratio’ and ‘average stretch’ among serviced measurement requests. Further, by sampling experiments on real-network perfSONAR measurement data sets, we show that our SPS algorithm successfully mitigates oversampling and further improves the satisfaction ratio. Our SPS scheme and evaluation results are vital to manage large-scale measurement infrastructures used for meeting monitoring objectives in the next-generation applications and networks.     

RRR: recursive round robin scheduler

Scheduling has been an interesting problem since its inception. In the context of real-time networks, a scheduling algorithm is concerned with dispatching streams of packets sharing the same bandwidth such that certain guaranteed performance for each stream like rate and delay bound is provided. This function has a wide range of applications in network elements such as host adaptors, routers and switches. This paper proposes and describes a new scheduling algorithm named as recursive round robin (RRR) scheduler. It is based on the concept of the construction of a scheduling tree in which distinct cell streams are scheduled recursively. Special emphasis is placed on the design and analysis of the scheduler. A delay bound is analytically derived for the scheduler and verified using simulation. The scheduler can work in either a work-conserving mode or non-work-conserving mode. It is shown that the work conserving scheduler is fair. Fairness indexes for the work conserving scheduler are analytically derived. The simple nature of this algorithm makes it possible to implement it at very high speeds, while considerably reducing the implementation cost.