数据中心中DVFS对程序性能影响模型的设计

数据中心以可接受的成本承载着超大规模的互联网应用.数据中心的能源消耗直接影响着数据中心的一次性建造成本和长期维护成本,是数据中心总体持有成本的重要组成部分.现代的数据中心普遍采用DVFS（Dynamic Voltage Frequency Scaling,动态电压频率调节）来提升单节点的能耗表现.但是,DVFS这一类机制同时影响应用的能源消耗和性能,而这一问题尚未被深入探索.本文专注于DVFS机制对应用程序性能的影响,提出了一个分析模型用来量化地刻画应用程序的性能同处理器频率之间的关系,可以预测程序在任意频率下的性能.具体来说,依据执行时访问内存子系统资源的不同,本文把程序的指令为两部分：片上指令和片外指令,并分别独立建模.片上指令指仅需访问片上资源就可以完成执行的指令,其执行时间同处理器频率成线性关系;片外指令指需要访问主存的指令,其执行时间同处理器频率无关.通过上述划分和对每部分执行时间的分别建模,我们可以获得应用程序的执行时间同处理器频率之间的量化模型.我们使用两个不同的平台和SPEC 2006中的所有标准程序验证该模型,平均误差不超过1.34%.     

一种面向多核系统的并行计算任务分配方法

随着多核处理器的普及,目前的大规模并行处理系统普遍采用多核处理器,这对于资源管理和调度提出了更高的要求.提出了基于共享Cache资源划分的方法,建立了面向多核处理器支持Cache资源分配的进程调度模型,设计并实现了并行任务到多核处理器的映射算法,更好地解决了大规模资源管理系统中面向多核处理器的任务分配问题,降低了使用共享Cache的多个进程运行时的相互干扰,提升了应用程序性能.     

多核处理器共享资源管理技术研究

现代多核处理器结构的设计使得集成在同一块芯片上的多个执行核共享各种硬件资源,如片上最后一级Cache、内存控制器、前端总线以及硬件预取单元等,而多线程的并行执行导致核与核之间其享资源的争用,造成系统整体性能的下降,如何有效地解决多核共享资源冲突来提升系统的整体性能以及应用程序的服务质量成为当今研究的热点.文章首先概要介...     

面向云计算数据中心的能耗建模方法

云计算对计算能力的需求,促进了大规模数据中心的飞速发展.与此同时,云计算数据中心产生了巨大的能耗.由于云计算的弹性服务和可扩展性等特性,云计算数据中心的硬件规模近年来极度膨胀,这使得过去分散的能耗问题变成了集中的能耗问题.因此,深入研究云计算数据中心的节能问题具有重要意义.为此,针对云计算数据中心的能耗问题,提出了一种精确度高的能耗模型来预测云计算数据中心单台服务器的能耗状况.精确的能量模型是很多能耗感知资源调度方法的研究基础,在大多数现有的云计算能耗研究中,多采用线性模型来描述能耗和资源利用率之间的关系.然而随着云计算数据中心服务器体系结构的变化,能耗和资源使用率的关系已经难以用简单的线性函数来描述.因此,从处理器性能计数器和系统使用情况入手,结合多元线性回归和非线性回归的数学方法,分析总结了不同参数和方法对服务器能耗建模的影响,提出了适合云计算数据中心基础架构的服务器能耗模型.实验结果表明,该能耗模型在只监控系统使用率的情况下,在系统稳定后,能耗预测精度可达到95%以上.     

多核处理器并行编程模型的研究与设计

为了在多核处理器上充分利用多核资源以提升程序性能,研究了多核处理器的体系结构和多核环境下可能影响并行程序性能的因素,实现了基于任务的并行编程模型.该模型提供了单任务数据并行和多任务并行两种并行处理方式,其中单任务数据并行使用cache块技术划分数据集,多任务并行使用任务密取的任务调度策略.用该模型实现了计算斐波那契数列的递归算法,实验结果表明,使用该模型编写多核并行程序可以达到较高的相对于串行计算的加速比.     

一种适应多核处理器核间通信机制的设计

随着单芯片上集成处理器内核数量的增加,在支持多核处理器的应用程序方面,核间通信变得更加重要.通过分析多核运行任务特点,根据处理核上运行任务功能的不同,将处理核分成两类:控制核和计算核.根据对核的分类,提出了一种新的核间通信模型,该模型提供了三种不同的通信通道.运用这三条通道,把应用程序的I/O部分从计算核迁移到控制核来提高多核的利用率,实验结果表明该方式有效提高核间协作以及核间通信的效率,提升处理器的利用率.     

基于多核处理器的VTD-XML节点查询执行性能优化

针对目前主流的多核处理器,研究了基于VTD-XML的节点查询执行性能优化,即基于预读策略从多线程并发执行和提高线程内存访问性能两个方面优化XML节点查询的性能。实验结果表明,提出的多线程XML文档解析框架可以充分利用多核处理器的计算资源,并有效地提高线程的内存访问性能,大大提高了XML节点查询的性能。     

基于多核处理器的VTD-XML节点查询执行性能优化

针对目前主流的多核处理器,研究了基于VTD-XML的节点查询执行性能优化,即基于预读策略从多线程并发执行和提高线程内存访问性能两个方面优化XML节点查询的性能。实验结果表明,提出的多线程XML文档解析框架可以充分利用多核处理器的计算资源,并有效地提高线程的内存访问性能,大大提高了XML节点查询的性能。     

异构处理器多操作系统协同技术研究

随着嵌入式设备应用场景日趋复杂的变化,异构多核架构逐渐成为嵌入式处理器的主流架构.目前,多核处理器主要采用的单操作系统模式在实际应用中存在诸多局限性.为了充分发挥异构处理器的多核特性,针对异构处理器不同核部署相应的操作系统并实现多操作系统协同处理技术至关重要.本文对异构多核处理器（ARM+DSP）操作系统进行了研究,在异构多核平台上成功移植了嵌入式Linux和国产DSP实时操作系统ReWorks;为实现ReWorks与Linux操作系统协同处理,本文对核间通信的关键技术进行分析研究,并以TI公司的AM5718为例,设计了一系列多核异构通信组件.经测试,本文设计的异构通信组件实现了在ARM上对DSP核进行ReWorks操作系统和应用程序的动态加载、Linux与ReWorks核间消息收发、以及Linux与ReWorks的协同计算等功能.     

Mbalancer:虚拟机内存资源动态预测与调配

在现代数据中心,虚拟化技术在资源管理、服务器整合、提高资源利用率等方面发挥了巨大的作用,已成为云计算架构中关键的抽象层次和重要的支撑性技术。在虚拟化环境中,如果要保证高资源利用率和系统性能,必须有一个高效的内存管理方法,使得虚拟机的物理内存大小能够满足应用程序不断变化的内存需求。因此,如何在单机以及数据中心内进行内存资源的动态调控,就成为了一个关键性问题。实现了一个低开销、高精确度的内存工作集跟踪机制,进而进行相应的本地或者全局的内存调控。采用了多种动态内存调控技术：气球技术能够在单机内有效地为各个虚拟机动态调节内存;远程缓存技术可在物理机之间进行内存调度;虚拟机迁移可将虚拟机负载在多个物理主机间进行均衡。深入分析了以上各种方案的优缺点,并根据内存超载的情况有针对性地设计了相应的调控策略,实验数据表明：所提出的预测式的内存资源管理方法能够对内存资源进行在线监控和动态调配,并有效地提高了数据中心的内存资源利用率,降低了数据中心能耗。     

Virtual machine consolidation based on interference modeling

Server consolidation is very attractive for cloud computing platforms to improve energy efficiency and resource utilization. Advances in multi-core processors and virtualization technologies have enabled many workloads to be consolidated in a physical server. However, current virtualization technologies do not ensure performance isolation among guest virtual machines, which results in degraded performance due to contention in shared resources along with violation of service level agreement (SLA) of the cloud service. In that sense, minimizing performance interference among co-located virtual machines is the key factor of successful server consolidation policy in the cloud computing platforms. In this work, we propose a performance model that considers interferences in the shared last-level cache and memory bus. Our performance interference model can estimate how much an application will hurt others and how much an application will suffer from others. We also present a virtual machine consolidation method called swim which is based on our interference model. Experimental results show that the average performance degradation ratio by swim is comparable to the optimal allocation.     

一种基于硬件计数器的虚拟机性能干扰估算方法

虚拟化技术已成为云计算平台中的关键性支撑技术.它极大地提高了数据中心的资源利用率,降低了管理成本和能源消耗,但同时也为数据中心带来了新的问题——性能干扰.同一平台上的多虚拟机过度竞争某一底层硬件资源(如CPU,Cache等),会造成虚拟机性能严重下降;而出于安全性和可移植性的考虑,底层平台管理者需要尽量避免侵入式监测上层虚拟机,因而,如何透明而有效地从底层估算虚拟机性能干扰,成为虚拟化平台管理者必须面临的一个挑战.为应对以上挑战,提出了一种基于硬件计数器的虚拟机性能干扰估算方法.硬件计数器是程序运行期间产生的硬件事件信息(如CPU时间片、缓存失效次数等),已有工作主要利用大规模分布式系统任务相似性查找产生异常硬件计数器数据的节点,而没有探究硬件事件变化与性能干扰之间的直接关系.通过实验研究发现,硬件计数器(last level cache misses rates,简称LLC misses rates)与不同资源需求的应用性能干扰存在不同的关联关系;以此建立虚拟机性能干扰估算模型,估算虚拟机性能.实验结果表明:该方法可以有效地预测CPU密集型应用和网络密集型应用的性能干扰大小,并仅为系统带来小于10%的开销.     

Dynamic forecast scheduling algorithm for virtual machine placement in cloud computing environment

In most cloud computing platforms, the virtual machine quotas are seldom changed once initialized, although the current allocated resources are not efficiently utilized. The average utilization of cloud servers in most datacenters can be improved through virtual machine placement optimization. How to dynamically forecast the resource usage becomes a key problem. This paper proposes a scheduling algorithm called virtual machine dynamic forecast scheduling (VM-DFS) to deploy virtual machines in a cloud computing environment. In this algorithm, through analysis of historical memory consumption, the most suitable physical machine can be selected to place a virtual machine according to future consumption forecast. This paper formalizes the virtual machine placement problem as a bin-packing problem, which can be solved by the first-fit decreasing scheme. Through this method, for specific virtual machine requirements of applications, we can minimize the number of physical machines. The VM-DFS algorithm is verified through the CloudSim simulator. Our experiments are carried out on different numbers of virtual machine requests. Through analysis of the experimental results, we find that VM-DFS can save 17.08 % physical machines on the average, which outperforms most of the state-of-the-art systems.     

Dynamic I/O-Aware Scheduling for Batch-Mode Applications on Chip Multiprocessor Systems of Cluster Platforms简

Efficiency of batch processing is becoming increasingly important for many modern commercial service centers, e.g., clusters and cloud computing datacenters. However, periodical resource contentions have become the major performance obstacles for concurrently running applications on mainstream CMP servers. I/O contention is such a kind of obstacle, which may impede both the co-running performance of batch jobs and the system throughput seriously. In this paper, a dynamic I/O-aware scheduling algorithm is proposed to lower the impacts of I/O contention and to enhance the co-running performance in batch processing. We set up our environment on an 8-socket, 64-core server in Dawning Linux Cluster. Fifteen workloads ranging from 8 jobs to 256 jobs are evaluated. Our experimental results show significant improvements on the throughputs of the workloads, which range from 7% to 431%. Meanwhile, noticeable improvements on the slowdown of workloads and the average runtime for each job can be achieved. These results show that a well-tuned dynamic I/O-aware scheduler is beneficial for batch-mode services. It can also enhance the resource utilization via throughput improvement on modern service platforms.     

A hybrid cloud controller for vertical memory elasticity: A control-theoretic approach

Web-facing applications are expected to provide certain performance guarantees despite dynamic and continuous workload changes. As a result, application owners are using cloud computing as it offers the ability to dynamically provision computing resources (e.g., memory, CPU) in response to changes in workload demands to meet performance targets and eliminates upfront costs. Horizontal, vertical, and the combination of the two are the possible dimensions that cloud application can be scaled in terms of the allocated resources. In vertical elasticity as the focus of this work, the size of virtual machines (VMs) can be adjusted in terms of allocated computing resources according to the runtime workload. A commonly used vertical resource elasticity approach is realized by deciding based on resource utilization, named capacity-based. While a new trend is to use the application performance as a decision making criterion, and such an approach is named performance-based. This paper discusses these two approaches and proposes a novel hybrid elasticity approach that takes into account both the application performance and the resource utilization to leverage the benefits of both approaches. The proposed approach is used in realizing vertical elasticity of memory (named as vertical memory elasticity), where the allocated memory of the VM is auto-scaled at runtime. To this aim, we use control theory to synthesize a feedback controller that meets the application performance constraints by auto-scaling the allocated memory, i.e., applying vertical memory elasticity. Different from the existing vertical resource elasticity approaches, the novelty of our work lies in utilizing both the memory utilization and application response time as decision making criteria. To verify the resource efficiency and the ability of the controller in handling unexpected workloads, we have implemented the controller on top of the Xen hypervisor and performed a series of experiments using the RUBBoS interactive benchmark application, under synthetic and real workloads including Wikipedia and FIFA. The results reveal that the hybrid controller meets the application performance target with better performance stability (i.e., lower standard deviation of response time), while achieving a high memory utilization (close to 83%), and allocating less memory compared to all other baseline controllers.     

面向高性能计算机的海量数据处理平台实现与评测

高性能计算机主要应用于传统的科学计算领域,而在云计算时代,数据密集型应用成为一大类新型应用,已经变得越来越重要.主要探索如何在高性能计算机上高效地进行海量数据处理,使高性能计算机在进行科学计算的同时,能够非常好地支持数据密集型应用,拓展高性能计算机的应用领域.分析了高性能计算机上MapReduce模型实现和部署的可行性之后,在高性能计算环境中进行了实验.实验结果表明,存储系统的并行I/O能力不能充分发挥,是造成系统无法高效运行的主要瓶颈.而导致这个性能瓶颈的原因,是高并发带来的对集群文件系统资源的竞争和冲突.最后,提出了几种解决集群文件系统资源冲突的方案,这是今后的研究方向.     

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.     

一种共享主存二维SIMD结构资源分配算法的改进与实现

共享主存二维SIMD结构已经广泛应用于多媒体处理加速部件,其数据并行性可以大大提高处理器的运算能力。目前,已有一些针对共享主存二维SIMD结构编译优化方面的研究,这些编译优化技术能有效地提高各种多媒体应用程序的加速比。但是,分析可知,这些优化方法的平均资源利用率只有约50％。本文基于对多媒体应用程序在共享主存二维 维SIMD结构上的执行过程分析,根据原有算法并适当修改经典图着色寄存器分配算法,提出了一种改进的资源分的目的。实验结果说明,该算法的改进对于大部分多媒体应用程序的性能有显著的提高。     

Optimization methodology of dynamic data structures based on genetic algorithms for multimedia embedded systems

Modern multimedia application exhibit high resource utilization. In order to efficiently run this kind of applications in embedded systems, the dynamic memory subsystem needs to be optimized. A key role in this optimization is played by the dynamic data structures that reside in every real-life application. This paper presents a novel and automated way to optimize dynamic data structures. The search space is pruned using genetic algorithms that converge to the best multilayered data structure implementation for the targeted applications.     

Resource management framework for collaborative computing systems over multiple virtual machines

A resource management framework for collaborative computing systems over multiple virtual machines (CCSMVM) is presented to increase the performance of computing systems by improving the resource utilization, which has constructed a scalable computing environment for resource on-demand utilization. We design a resource management framework based on the advantages of some components in grid computing platform, virtualized platform and cloud computing platform to reduce computing systems overheads and maintain workloads balancing with the supporting of virtual appliance, Xen API, applications virtualization and so on. The content of collaborate computing, the basis of virtualized resource management and some key technologies including resource planning, resource allocation, resource adjustment and resource release and collaborative computing scheduling are designed in detail. A prototype is designed, and some experiments have verified the correctness and feasibility of our prototype. System evaluations show that the time in resource allocation and resource release is proportional to the quantity of virtual machines, but not the time in the virtual machines migrations. CCSMVM has higher CPU utilization and better performance than other systems, such as Eucalyptus 2.0, Globus4.0, et al. It is concluded that CCSMVM can accelerate the execution of systems by improving average CPU utilization from the results of comparative analysis with other systems, so it is better than others. Our study on resource management framework has some significance to the optimization of the performance in virtual computing systems.