基于Xen的虚拟磁盘调度算法改进

Xen目前所采用的I/O调度算法能够较好的保证公平性;但在实际应用中,不同的虚拟机可能有不同的性能需求。该文研究了Xen虚拟存储的实现和I/O调度算法的原理,提出了基于反馈的动态优先级调度策略,通过对比测试验证了磁盘带宽在不同虚拟机之间的按需分配。     

一种动态优先级排序的虚拟机I/O调度算法

I/O任务调度是影响I/O密集型虚拟机性能的重要因素。现有调度方法主要是针对虚拟机整机I/O带宽的优化,较少兼顾各虚拟域与全局性能,也无法满足域间差异化服务的要求。针对现有方法的不足,提出了一种动态优先级排序的虚拟机I/O调度算法DPS。该算法基于多属性决策理论,以离差最大化方法计算I/O任务的优先级评估属性权重,对I/O任务优先级进行综合评估;通过引入任务所在虚拟域价值,体现云计算环境下虚拟域重要性差异。在Xen系统中通过实验评测DPS调度虚拟化网卡的性能,结果表明,DPS能够有效提高指定域与全局的I/O任务截止期保证率、整机I/O带宽,并能为不同虚拟域的I/O应用提供差异化服务。     

一种基于虚拟机的高效磁盘I/O特征分析方法

于磁盘系统的机械运动本质,磁盘系统I/O往往会成为计算机系统的性能瓶颈.为了有效地提高系统性能,收集和分析应用系统的磁盘I/O特征信息将成为性能优化工作的重要基础.与以往I/O特征分析方法不同,给出了一种基于Xen 3.0虚拟机系统的磁盘I/O特征在线分析方法.在虚拟机环境下,该磁盘I/O特征采集方法可以透明地应用于任意无须修改的操作系统.该方法可以高效地在线采集多种基本I/O特征数据,其中包括:磁盘I/O块大小、I/O延迟、I/O时间间隔、I/O空间局部性、时间局部性以及磁盘I/O操作热点分布.通过测试和分析,该在线I/O分析方法有着较小的系统开销,并且对应用系统I/O性能的影响很小.此外,还给出了在大文件拷贝、基于Filebench的filemirco和varmail等工作负载下的I/O特征分析结果.     

虚拟化环境中线程级SLO保障的I/O软件框架研究

云计算产业的快速发展使得虚拟化技术在各大云服务商心目中占据重要地位。为了获取更高的利润,云服务商需要在保障用户体验的前提下尽可能地利用设备性能。通过利用I/O请求的优先级和重要性等信息,研究者们已经在Linux内核中实现了很多提高程序性能的方法。然而,虚拟机中的这些信息在传递到宿主机的过程中会丢失,所以 提出了一种基于服务水平目标SLO的I/O保障框架。首先分析了I/O请求优先级等信息丢失的原因,并提出了传递这些信息需要解决的关键性问题。在此基础上,本文提出的框架通过对Linux内核、virtio协议以及KVM的I/O虚拟化程序QEMU进行扩展,成功地将虚拟机线程的SLO信息传送至宿主机并在此基础上实现了基于SLO信息的调度器。最后,通过实验验证了框架的可行性,优先级最高的线程吞吐量可以达到260 KB/s,优先级最低的线程吞吐量只有10 KB/s,成功证明了由框架传递下来的SLO信息对宿主机中调度器的调度起到了积极作用。     

Adaptive cycle management in soft real-time disk retrieval

The objective of this study is to determine the right cycle management policy to service periodic soft real-time disk retrieval. Cycle-based disk scheduling provides an effective way of exploiting the disk bandwidth and meeting the soft real-time requirements of individual I/O requests. It is widely used in real-time retrieval of multimedia data blocks. Interestingly, the issue of cycle management with respect to dynamically changing workloads has not been receiving proper attention despite its significant engineering implications on the system behavior. When cycle length remains constant regardless of varying I/O workload intensity, it may cause under-utilization of disk bandwidth capacity or unnecessarily long service startup latency. In this work, we present a novel cycle management policy which dynamically adapts to the varying workload. We develop pre-buffering policy which makes the adaptive cycle management policy robust against starvation. The proposed approach elaborately determines the cycle length and the respective buffer size for pre-buffering. Performance study reveals a number of valuable observations. Adaptive cycle length management with incremental pre-buffering exhibits superior performance to the other cycle management policies in startup latency, jitter and buffer requirement. It is found that servicing low playback rate contents such as video contents for 3G cellular network requires rather different treatment in disk subsystem capacity planning and call admission criteria because relatively significant fraction of I/O latency is taken up by plain disk overhead.     

Providing QOS guarantees for disk I/O

In this paper, we address the problem of providing different levels of performance guarantees or quality of service for disk I/O. We classify disk requests into three categories based on the provided level of service. We propose an integrated scheme that provides different levels of performance guarantees in a single system. We propose and evaluate a mechanism for providing deterministic service for variable-bit-rate streams at the disk. We will show that, through proper admission control and bandwidth allocation, requests in different categories can be ensured of performance guarantees without getting impacted by requests in other categories. We evaluate the impact of scheduling policy decisions on the provided service. We also quantify the improvements in stream throughput possible by using statistical guarantees instead of deterministic guarantees in the context of the proposed approach.     

高强度I/O的应用对并行存储系统的挑战和解决方法研究

具有高I/O密集特性的高性能计算应用对高性能计算机存储系统综合性能的要求越来越高.以石油地震勘探数据处理为代表的一类重要应用表现出I/O数据量巨大、I/O访问密度大,对单个磁盘阵列存储部件的读写带宽要求高的特征.在Lustre文件系统中,充当对象存储服务功能的磁盘阵列设备输出带宽的不足将成为阻碍存储系统整体性能发挥的重要因素.针对此类问题,提出了一种缓存管理方法,分别在客户端添加VDISK模块,在OST端添加Cache模块,二者协同提高并行文件系统I/O的输出带宽的使用效率;另外,充分利用客户端空闲内存以及客户端之间的通信带宽,降低应用程序对磁盘阵列设备输出带宽的要求.通过大规模并行模型的验证表明,VDISK提高了实际可用的输出带宽,提高了外部存储系统的I/O效率.     

基于重复数据删除的虚拟桌面存储优化技术

虚拟桌面基础架构依靠数据中心海量的云基础设施,为用户按需提供虚拟桌面部署所需的软硬件资源,但同时面临存储资源利用率低和虚拟机启动慢的困境.针对虚拟桌面存储中具有大量数据冗余的特性,采用重复数据删除技术缩减虚拟桌面基础架构的存储空间需求;并利用服务器本地磁盘缓存以及共享存储池内的固态硬盘来优化虚拟机的启动性能.通过原型实现,发现相比于基于内容分块的策略,静态分块策略更适合虚拟桌面存储进行重复数据删除,最优的分块大小为4KB,并能够缩减85%的存储空间容量;通过服务器本地磁盘缓存和基于闪存的固态硬盘进行I/O优化,虚拟机的启动速度能够获得35%的提升.     

一种高效的虚拟机快照管理系统

链式虚拟磁盘快照技术是广泛实现的一种支持虚拟机失效恢复的技术手段,针对链式结构磁盘快照技术引入多余I／O请求导致虚拟机性能低效问题,分析和研究了支持高效虚拟机快照实现的无链式关联磁盘快照技术,通过集成ZFS及OCFS2文件系统给出了无链式关联磁盘快照技术实现,并设计和实现了虚拟机快照存储组织模型和快速检索算法,提高了虚拟机快照检索效率．设计试验验证了系统的有效性．     

基于微内核的虚拟机I/O安全机制

NOVA等微内核虚拟化架构解决了宏内核平台可信计算基体积和攻击面过大的问题, 但其仍缺乏虚拟机分等级保护和I/O资源访问控制等安全机制. 本文提出了安全域的概念, 并将虚拟机划分至不同的安全域, 进而建立可定制的I/O资源访问控制机制. 通过将访问控制模块添加至I/O资源访问的关键代码路径, 实现了不同安全域的I/O资源访问控制. 实验表明, 该机制提高了数据的隔离性与安全性, 仅对计算密集型、I/O密集型任务造成了较小的性能损耗.     

An interleaving crescent broadcasting protocol for near video-on-demand services

This work presents a novel interleaving crescent broadcasting protocol for near video-on-demand service. The interleaving crescent broadcasting protocol is a trade-off among the subscriber’s access latency, maximum buffer requirement, needed subscriber’s bandwidth, and maximum disk I/O transfer rate. A longer subscriber’s access latency may cause a subscriber to leave. A lower maximum buffer requirement, a lower needed subscriber’s bandwidth, and a lower maximum disk I/O transfer rate reduce subscribers’ costs. The interleaving crescent broadcasting protocol not only makes access latency shorter, but also lowers the overall system’s cost. We prove the correctness of the interleaving crescent protocol; provide mathematical analyses to demonstrate its efficiency.     

面向应用服务级目标的虚拟化资源管理

在虚拟环境中实现应用服务级目标,是当前数据中心系统管理的关键问题之一.解决该问题有两个方面的要求:一方面,在虚拟化层次和范围内,能够动态和分布式地按需调整虚拟机资源分配;另一方面,在虚拟化范围之外,能够控制由于虚拟机对非虚拟化资源的竞争所导致的性能干扰,实现虚拟机性能隔离.然而,已有工作不适用于虚拟化数据中心场景.提出一种面向应用服务级目标的虚拟化资源管理方法.首先,该方法基于反馈控制理论,通过动态调整虚拟机资源分配来实现每个应用的服务器目标;同时,还设计了一个两层结构的自适应机制,使得应用模型能够动态地捕捉虚拟机资源分配与应用性能的时变非线性关系;最后,该方法通过仲裁不同应用的资源分配请求来控制虚拟机在非虚拟化资源上的竞争干扰.实验在基于Xen的机群环境中检验了该方法在RUBiS系统和TPC-W基准上的效果.实验结果显示,该方法的应用服务级目标实现率比两种对比方法平均高29.2%,而应用服务级目标平均偏离率比它们平均低50.1%.另一方面,当RUBiS系统和TPC-W基准竞争非虚拟化的磁盘I/O资源时,该方法通过抑制TPC-W基准28.7%的处理器资源需求来优先满足RUBiS系统的磁盘I/O需求.     

I/O-aware bandwidth allocation for petascale computing systems

In the Big Data era, the gap between the storage performance and an application’s I/O requirement is increasing. I/O congestion caused by concurrent storage accesses from multiple applications is inevitable and severely harms the performance. Conventional approaches either focus on optimizing an application’s access pattern individually or handle I/O requests on a low-level storage layer without any knowledge from the upper-level applications. In this paper, we present a novel I/O-aware bandwidth allocation framework to coordinate ongoing I/O requests on petascale computing systems. The motivation behind this innovation is that the resource management system has a holistic view of both the system state and jobs’ activities and can dynamically control the jobs’ status or allocate resource on the fly during their execution. We treat a job’s I/O requests as periodical sub-jobs within its lifecycle and transform the I/O congestion issue into a classical scheduling problem. Based on this model, we propose a bandwidth management mechanism as an extension to the existing scheduling system. We design several bandwidth allocation policies with different optimization objectives either on user-oriented metrics or system performance. We conduct extensive trace-based simulations using real job traces and I/O traces from a production IBM Blue Gene/Q system at Argonne National Laboratory. Experimental results demonstrate that our new design can improve job performance by more than 30%, as well as increasing system performance.     

面向虚拟机的远程磁盘缓存

在虚拟机(virtual machine)系统中,随着虚拟机数量和应用程序需求的不断增长,内存容量已经成为应用程序性能的主要瓶颈。为了提升内存密集型和I/O密集型程序的页面交换性能,提出了虚拟机的远程磁盘缓存机制REMOCA,它允许运行在一台物理主机上的虚拟机将其他物理主机的内存作为其二级磁盘缓存。由于网络传输延迟远远小于磁盘访问,用网络传输代替磁盘访问就能够有效地降低虚拟机的平均磁盘访问延迟。REMOCA的目标就要尽可能地减少磁盘访问。REMOCA运行在虚拟机管理器中,其基本工作原理是截获并处理虚拟机的页面淘汰、磁盘访问等事件。REMOCA能够与现有的虚拟机内存管理机制(如气球技术、影子缓存)相结合,从而提供更加灵活的内存资源管理策略。实验数据表明,REMOCA能有效地降低页面抖动对虚拟机性能的影响,并在很大程度上提升虚拟机中I/O密集型应用的性能。     

Reducing I/O variability using dynamic I/O path characterization in petascale storage systems

In petascale systems with a million CPU cores, scalable and consistent I/O performance is becoming increasingly difficult to sustain mainly because of I/O variability. The I/O variability is caused by concurrently running processes/jobs competing for I/O or a RAID rebuild when a disk drive fails. We present a mechanism that stripes across a selected subset of I/O nodes with the lightest workload at runtime to achieve the highest I/O bandwidth available in the system. In this paper, we propose a probing mechanism to enable application-level dynamic file striping to mitigate I/O variability. We implement the proposed mechanism in the high-level I/O library that enables memory-to-file data layout transformation and allows transparent file partitioning using subfiling. Subfiling is a technique that partitions data into a set of files of smaller size and manages file access to them, making data to be treated as a single, normal file to users. We demonstrate that our bandwidth probing mechanism can successfully identify temporally slower I/O nodes without noticeable runtime overhead. Experimental results on NERSC’s systems also show that our approach isolates I/O variability effectively on shared systems and improves overall collective I/O performance with less variation.     

Fault-tolerant feedback virtual machine deployment based on user-personalized requirements

A key requirement of the cloud platform is the reasonable deployment of its large-scale virtual machine infrastructure. The mapping relation between the virtual node and the physical node determines the specific resource distribution strategy and reliability of the virtual machine deployment. Resource distribution strategy has an important effect on performance, energy consumption, and guarantee of the quality of service of the computer, and serves an important role in the deployment of the virtual machine. To solve the problem of meeting the fault-tolerance requirement and guarantee high reliability of the application system based on the full use of the cloud resource under the prerequisite of various demands, the deployment framework of the feedback virtual machine in cloud platform facing the individual user’s demands of fault-tolerance level and the corresponding deployment algorithm of the virtual machine are proposed in this paper. Resource distribution strategy can deploy the virtual machine in the physical nodes where the resource is mutually complementary according to the users’ different requirements on virtual resources. The deployment framework of the virtual machine in this paper can provide a reliable computer configuration according to the specific fault-tolerance requirements of the user while considering the usage rate of the physical resources of the cloud platform. The experimental result shows that the method proposed in this paper can provide flexible and reliable select permission of fault-tolerance level to the user in the virtual machine deployment process, provide a pertinent individual fault-tolerant deployment method of the virtual machine to the user, and guarantee to meet the user service in a large probability to some extent.     

Efficient support for interactive service in multi-resolution VOD systems

Advances in high-speed networks and multimedia technologies have made it feasible to provide video-on-demand (VOD) services to users. However, it is still a challenging task to design a cost-effective VOD system that can support a large number of clients (who may have different quality of service (QoS) requirements) and, at the same time, provide different types of VCR functionalities. Although it has been recognized that VCR operations are important functionalities in providing VOD service, techniques proposed in the past for providing VCR operations may require additional system resources, such as extra disk I/O, additional buffer space, as well as network bandwidth. In this paper, we consider the design of a VOD storage server that has the following features: (1) provision of different levels of display resolutions to users who have different QoS requirements, (2) provision of different types of VCR functionalities, such as fast forward and rewind, without imposing additional demand on the system buffer space, I/O bandwidth, and network bandwidth, and (3) guarantees of the load-balancing property across all disks during normal and VCR display periods. The above-mentioned features are especially important because they simplify the design of the buffer space, I/O, and network resource allocation policies of the VOD storage system. The load-balancing property also ensures that no single disk will be the bottleneck of the system. In this paper, we propose data block placement, admission control, and I/O-scheduling algorithms, as well as determine the corresponding buffer space requirements of the proposed VOD storage system. We show that the proposed VOD system can provide VCR and multi-resolution services to the viewing clients and at the same time maintain the load-balancing property. Received June 9, 1998 / Accepted April 26, 1999     

Transparently bridging semantic gap in CPU management for virtualized environments

Consolidated environments are progressively accommodating diverse and unpredictable workloads in conjunction with virtual desktop infrastructure and cloud computing. Unpredictable workloads, however, aggravate the semantic gap between the virtual machine monitor and guest operating systems, leading to inefficient resource management. In particular, CPU management for virtual machines has a critical impact on I/O performance in cases where the virtual machine monitor is agnostic about the internal workloads of each virtual machine. This paper presents virtual machine scheduling techniques for transparently bridging the semantic gap that is a result of consolidated workloads. To enable us to achieve this goal, we ensure that the virtual machine monitor is aware of task-level I/O-boundedness inside a virtual machine using inference techniques, thereby improving I/O performance without compromising CPU fairness. In addition, we address performance anomalies arising from the indirect use of I/O devices via a driver virtual machine at the scheduling level. The proposed techniques are implemented on the Xen virtual machine monitor and evaluated with micro-benchmarks and real workloads on Linux and Windows guest operating systems.     

VM3: Measuring,modeling and managing VM shared resources

With cloud and utility computing models gaining significant momentum, data centers are increasingly employing virtualization and consolidation as a means to support a large number of disparate applications running simultaneously on a chip-multiprocessor (CMP) server. In such environments, contention for shared platform resources (CPU cores, shared cache space, shared memory bandwidth, etc.) can have a significant effect on each virtual machine’s performance. In this paper, we investigate the shared resource contention problem for virtual machines by: (a) measuring the effects of shared platform resources on virtual machine performance, (b) proposing a model for estimating shared resource contention effects, and (c) proposing a transition from a virtual machine (VM) to a virtual platform architecture (VPA) that enables transparent shared resource management through architectural mechanisms for monitoring and enforcement. Our measurement and modeling experiments are based on a consolidation benchmark (vConsolidate) running on a state-of-the-art CMP server. Our virtual platform architecture experiments are based on detailed simulations of consolidation scenarios. Through detailed measurements and simulations, we show that shared resource contention affects virtual machine performance significantly and emphasize that virtual platform architectures is a must for future virtualized datacenters.     

Resource policing to support fine-grain cycle stealing in networks of workstations

We present the design, implementation, and performance evaluation of a suite of resource policing mechanisms that allow guest processes to efficiently and unobtrusively exploit otherwise idle workstation resources. Unlike traditional policies that harvest cycles only from unused machines, we employ fine-grained cycle stealing to exploit resources even from machines that have active users. We developed a suite of kernel extensions that enable these policies to operate without significantly impacting host processes: 1) a new starvation-level CPU priority for guest jobs, 2) a new page replacement policy that imposes hard bounds on physical memory usage by guest processes, and 3) a new I/O scheduling mechanism called rate windows that throttle guest processes' usage of I/O and network bandwidth. We evaluate both the individual impacts of each mechanism, and their utility for our fine-grain cycle stealing.