构建度量区时序逻辑的时间自动机

在实时系统的形式验证中,为了直接验证带有明显时间约束的性质,选用了一种被广泛接受的(线性时间)实时时序逻辑——度量区时序逻辑来描述待验证的性质;提出了基于迁移的扩展时间B chi自动机;构建了度量区时序逻辑的基于迁移的扩展时间B chi自动机。这样扩展了已有实时系统模型检测工具的性质规范语言的表达能力,使其能直接处理和验证带有明显时间约束的性质。实现的工具表明,该算法有效且可行,并且显著地减少了结果自动机节点和迁移的数量,从而降低了结果自动机的大小,有利于进一步的模型检测过程。     

基于AB角理论的网格负载平衡研究

针对网格系统中的负载不平衡和负载迁移问题,提出利用AB角的方法来进行快速的负载迁移以实现负载平衡。分析了网格系统中的负载平衡问题,讨论了收集节点信息的和判断节点的状态的方法,并根据节点的状态设计了能够为它确定B节点的算法,以及系统中如何使用AB角理论对负载进行快速迁移。实验结果表明,该算法考虑到了网格资源的异构性和任务对资源类型需求的不同,从而缩短了任务的执行时间和获得了更高的迁移成功率。     

MOSIX集群系统中DFSA机制研究

MOSIX是基于Linux的集群计算系统,它通过抢占式进程迁移使负载在集群节点间达到均衡。DFSA(DirectFileSystemAccess)是一种能够提高集群文件系统性能的机制,它允许迁移的进程直接访问当前节点上的文件。MOSIX通过其文件系统MFS(MOSIXFileSystem)引入了对DFSA的支持,实现了把进程迁移到数据的策略。文中简要介绍了MOSIX系统,然后对DFSA机制进行了分析,最后研究了MOSIX中DFSA机制的实现。     

基于网格的分布式仿真系统负载平衡研究

对进程迁移的研究起源于分布式处理的出现，它使得分布式系统内动态负载平衡成为可能，并可显著提高系统的容错性和可用性；网格是新一代信息基础设施，其目的是实现资源共享利用网格提供的信息，将过载节点上的进程迁移到合适的其它节点上运行，就可以实现真正的负载平衡。该文首先简要介绍了进程迁移及网格的资源管理功能，根据现有的基于HLA构建的分布式仿真系统存在的缺陷，利用进程迁移、网格技术的优势研究并提出了分布式仿真系统负载平衡使能模块(LBEM)。现初步实现了网格环境中进程迁移的部分功能，该部分的研究工作是后续研究的基础。     

基于负载预测的虚拟机动态调度算法研究与实现

在云计算系统中为了实现负载均衡和资源的高效利用,需要在虚拟机粒度上对云计算系统进行调度,通过热迁移技术将虚拟机从高负载物理节点迁移到低负载物理节点。把负载预测技术和虚拟机动态调度技术相结合,提出了LFS算法,通过虚拟机历史负载数据对虚拟机未来的负载变化情况进行预测,然后根据预测结果对虚拟机进行调度,能够有效地避免云计算系统中高负载物理节点出现,实现负载均衡,提高资源使用率。     

MOSIX集群系统中DFSA机制研究

MOSIX是基于Linux的集群计算系统,它通过抢占式进程迁移使负载在集群节点间达到均衡.DFSA(Direct FileSystem Access)是一种能够提高集群文件系统性能的机制,它允许迁移的进程直接访问当前节点上的文件.MOSIX通过其文件系统MFS(MOSIX File System)引入了对DFSA的支持,实现了把进程迁移到数据的策略.文中简要介绍了MOSIX系统,然后对DFSA机制进行了分析,最后研究了MOSIX中DFSA机制的实现.     

面向云计算的多虚拟机管理模型的设计

提出基于P2P结构的多虚拟机管理模型,并实现其原型系统。采用P2P结构组织宿主机节点,利用组播实现资源发现;提出虚拟机动态迁移算法,自主触发节点间虚拟机动态迁移;利用选举的根节点映射云计算用户请求至宿主机,实现节点上虚拟机的按需创建、删除及停止等操作。实验表明:该模型具有收敛时间短、带宽占用率低及高可用性的特点,能够实现云计算资源的动态负载均衡。     

基于TL_Sheduling的异构集群负载均衡算法的研究与应用

异构环境下"集群系统"负载均衡技术的关键是实现进程的跨平台迁移,对于进程的跨平台迁移所带来的困难和开销是巨大的.在传统进程迁移算法基础上,充分考虑节点对提交作业的适应性,提出一种新的TL-Scheduling负载均衡算法,能够控制作业提交到适合它执行的节点上,该算法可以有效提高系统负载均衡和作业执行效率.在其基础上,研究设计了基于XML业务流程模板的卫星作业调度系统,使得作业调度系统更加实用化.     

基于反馈的实时集群系统中的负载平衡方法

为了提高实时集群系统中各节点的利用率并防止出现CPU使用率的热点,研究了实时集群系统中基于反馈控制的实时调度框架,并提出了一种新的面向任意图结构的负载平衡算法.该算法基于差异迁移系数和扩散负载平衡原理指导任务迁移以实现系统负载均衡,同时结合反馈控制以避免系统节点使用率振荡.实验结果表明,该算法不仅实现了节点间负载平衡,有效避免了局部热点,而且与反馈控制算法有机集成,保证整个系统稳定运行.     

虚拟化环境中服务监控调度系统的设计与实现

针对虚拟化环境中因服务访问量的不同导致个别节点负载过大的现象,设计并实现了虚拟化环境中服务的监控调度系统,该系统主要包括对虚拟机的监控,对部署在虚拟机上的Web Services服务的监控以及对服务的迁移和调度。本文搭建了虚拟化实验环境并设计了整个系统的功能模块,实现了在个别虚拟机过载时,通过上报Web Services服务信息至调度中心,由调度中心决定服务迁移的目的节点,进行服务的迁移调度。     

TYPHOON: mobile distributed hash tables

TYPHOON is a capability-aware peer-to-peer (P2P) system. It exploits the heterogeneity of nodes in the system based on the concept of virtual homes. Nodes participating in the system are classified as good and inactive. TYPHOON uses resources provided by good peers. It is thus more reliable and agile than a naive structured P2P system. When a good peer is overloaded, it picks a suitable inactive node and migrates some loads (i.e., virtual homes) to that node. However, migration of virtual homes may cause instability in the system. TYPHOON thus incorporates a mechanism for tracking virtual homes. A migrated home can receive states of relevant homes using an adaptive, logical tree structure that can also react to system heterogeneity, node loading and network locality. A migrated home can also proactively discover the state of an interested home to ensure the correctness of lookup. We evaluate TYPHOON using theoretical and simulation analysis. We also benchmark TYPHOON using a prototype system on 34 desktop PCs. The results all confirm the effectiveness of TYPHOON.     

基于Hadoop的访问热点副本迁移技术

提出一种云环境下的访问热点负载均衡模型：基于节点的吞吐量与响应时间等主要参考指标,构建节点负载判定模块;文件在HDFS存储的过程中,将文件对应的数据块编号与存储路径相结合,设计存放在数据节点中的数据块到文件目录映射表;提出一种基于节点负载以及节点的存储空间的迁移源节点和目标节点选择方法;基于机架感知的机制,制定一种动态副本迁移方案。最后利用执行器下发指令给相应的数据节点,执行具体的迁移任务以及完善迁移后副本因子等参数信息的调整。通过迅速扩散副本的方式,来增加热点文件的副本数量,使得系统能够对外提供更大的吞吐量,缩短系统反应时间。      

虚拟机迁移系统网络连接重定向技术*

针对虚拟机（virtual machine,VM）在不同子网间进行在线迁移时,因网络接入点发生改变而无法继续通信的问题,设计和实现了基于代理移动IP（proxy mobile IP,PMIP）的网络连接重定向系统。当VM迁移到外部子网时,将其看做移动节点,不改变IP地址,通过外部网络和家乡网络之间的双向隧道传输所有来自和发往该VM的网络数据。此方案中跨子网的在线迁移行为对VM以及与VM通信的节点透明。实验证明了该系统可以保证VM较快的网络切换速度和较好的网络通信性能。     

移动计算中维持可靠通讯链路的组件迁移

为了解决分布式系统中的动态负载均衡、容错、动态资源利用、隐式的颗粒度控制等问题,透明的组件迁移技术起了非常重要的作用。该文描述了一种在动态网络拓扑中有效地维持系统组件之间通讯链路的迁移机制,并考虑了对并发迁移情况的处理。提出的机制只用到较少的系统拓扑知识,在迁移过程中不阻塞进程,而只要求迁移组件本身停止计算,因而对非迁移系统组件的影响被减至最低。此迁移算法虽然是在移动代理系统中实现,但原理同样适用于其它分布式系统。     

A formal architectural model for logical agent mobility

The process of agent migration is the major difference between logical code mobility of software agents and physical mobility of mobile nodes in ad hoc networks. Without considering agent transfer, it would make little sense to mention the modeling of strong code mobility, which aims to make a migrated agent restarted exactly from the state when it was stopped before migration. From the perspective of system's architecture, this paper proposes a two-layer approach for the formal modeling of logical agent mobility (LAM) using predicate/transition (PrT) nets. We view a mobile agent system as a set of agent spaces and agents could migrate from one space to another. Each agent space is explicitly abstracted to be a component, consisting of an environmental part and an internal connector dynamically binding agents with their environment. We use a system net, agent nets, and a connector net to model the environment, agents, and the connector, respectively. In particular, agent nets are packed up as parts of tokens in system nets, so that agent transfer and location change are naturally captured by transition firing (token game) in Petri nets. Agent nets themselves are active only at specific places and disabled at all the other places in a system net. The semantics of such a two-layer LAM model is defined by transforming it into a PrT net. This facilitates the analysis of several properties about location, state, and connection. In addition, this paper also presents a case study of modeling and analyzing an information retrieval system with mobile agents.     

Memory-aware kernel mechanism and policies for improving internode load balancing on NUMA systems

Although nonuniform memory access architecture provides better scalability for multicore systems, cores accessing memory on remote nodes take longer than those accessing on local nodes. Remote memory access accompanied by contention for internode interconnection degrades performance. Properly mapping threads to cores and data accessed to their nodes can substantially improve performance and energy efficiency. However, an operating system kernel's load-balancing activity may migrate threads across nodes, which thus messes up the thread mapping. Besides, subsequent data mapping behavior pays for the cost of page migration to reduce remote memory access. Once unsuitable threads are migrated, it is detrimental to system performance. This paper focuses on improving the kernel's internode load balancing on nonuniform memory access systems. We develop a memory-aware kernel mechanism and policies to reduce remote memory access incurred by internode thread migration. The Linux kernel's load balancing mechanism is modified to incorporate selection policies in the internode thread migration, and the kernel is modified to track the amount of memory used by each thread on each node. With this information, well-designed policies can then choose suitable threads for internode migration. The purpose is to avoid migrating a thread that might incur relatively more remote memory access and page migration. The experimental results show that with our mechanism and the proposed selection policies, the system performance is substantially increased when compared with the unmodified Linux kernel that does not consider memory usage and always migrates the first-fit thread in the runqueue that can be migrated to the target central processing unit.     

基于负载特征的虚拟机迁移调度策略

为提高虚拟机迁移时的资源利用率及服务可用性,提出一种基于负载特征的虚拟机迁移调度策略。针对节点的触发类型和虚拟机的负载特征,采用多阈值方式触发迁移,完成对拟迁移的虚拟机以及迁移目标节点的选择。实验结果表明,该策略能够实现虚拟机迁移的自主管理,并能提升资源的使用效率,具有较好的自适应性。     

面向大规模数据接入系统的负载平衡机制

当前分布式系统负载平衡算法存在问题：1）算法建立的系统中各节点角色固定,系统不具有自适应性;2）算法的通用性不高;3）负载迁移任务巨大,且负载平衡周期过长等。针对这些问题,提出了混合式负载平衡算法。首先,设计了一个分布式系统接收模型。模型将系统任务分为三层：接收层、处理层和存储层。在接收层使用了自定义的通信协议提高系统的接收性能。然后,负载平衡算法采用随机负载迁移策略,根据系统中节点的负载状态,对负载任务进行随机迁移。通过这种策略解决负载平衡周期过长和负载回迁问题。最后,通过分布式控制节点选择策略,使系统中节点具有自适应性。实验结果显示,在百万数据源以下,系统各层平均延迟处于毫秒级,系统负载平衡平均耗时在3 min以下。实验证明了所提出的负载平衡机制具有周期短、任务响应迅速等特点,能够提高分布式系统的接收性能。     

Server-side QUIC connection migration to support microservice deployment at the edge

In edge computing environments, microservices are typically deployed in the form of containers. To maintain proximity between edge computing services and mobile users, containers need to be migrated between nodes. When migrating containers, however, it is important to consider that they typically have ongoing communications with client endpoints. Moreover, in case of connection-oriented protocols, communicating endpoints share a state (i.e., the connection), which needs to be migrated as well. Connection-oriented protocols like TCP were not designed having connection migration in mind, thus their connections cannot survive a change of IP address or port number. On the other hand, QUIC, a transport protocol recently standardised by IETF, provides a mechanism for client-side connection migration, whereas a server-side connection migration is not yet implemented nor investigated. In this work, we propose an extension of QUIC to support server-side connection migration when a container is migrated between servers. We designed three different strategies, fitting a diverse set of scenarios, wherein the migration procedure is either or not supported by a centralised entity, e.g., an orchestrator. We implemented and verified the proposed extension. Besides, we evaluated it on a real testbed, showing how each of the three strategies is impacted by different container migration techniques and container sizes. To conclude, we compared our solution against two alternatives based on TCP+DNS and MPTCP respectively, demonstrating performance improvements in terms service-migration time.