Efficient support for content‐aware request distribution and persistent connection in Web clusters

To support Web clusters with efficient dispatching mechanisms and policies, we propose a light‐weight TCP connection transfer mechanism, TCP Rebuilding, and use it to develop a content‐aware request dispatching platform, LVS‐CAD, in which the request dispatcher can extract and analyze the content in requests and then dispatch each request by its content or type of service requested. To efficiently support HTTP/1.1 persistent connection in Web clusters, request scheduling should be performed per request rather than per connection. Consequently, multiple TCP Rebuilding, as an extension to normal TCP Rebuilding, is proposed and implemented. On this platform, we also devise fast TCP module handshaking to process the handshaking between clients and the request dispatcher in the IP layer instead of in the TCP layer for faster response times. Furthermore, we also propose content‐aware request distribution policies that consider cache locality and various types of costs for dispatching requests in this platform, which makes the resourceutilization of Web servers more effective. Experimental results of a practical implementation on Linux show that the proposed system, mechanisms, and policies can effectively improve the performance of Web clusters. Copyright © 2007 John Wiley & Sons, Ltd.     

StopGap: elastic VMs to enhance server consolidation

Virtualized cloud infrastructures (also known as IaaS platforms) generally rely on a server consolidation system to pack virtual machines (VMs) on as few servers as possible. However, an important limitation of consolidation is not addressed by such systems. Because the managed VMs may be of various sizes (small, medium, large, etc.), VM packing may be obstructed when VMs do not fit available spaces. This phenomenon leaves servers with a set of unused resources (‘holes’). It is similar to memory fragmentation, a well‐known problem in operating system domain. In this paper, we propose a solution which consists in resizing VMs so that they can fit with holes. This operation leads to the management of what we call elastic VMs and requires cooperation between the application level and the IaaS level, because it impacts management at both levels. To this end, we propose a new resource negotiation and allocation model in the IaaS, called HRNM. We demonstrate HRNM's applicability through the implementation of a prototype compatible with two main IaaS managers (OpenStack and OpenNebula). By performing thorough experiments with SPECvirt_sc2010 (a reference benchmark for server consolidation), we show that the impact of HRNM on customer's application is negligible. Finally, using Google data center traces, we show an improvement of about 62.5% for the traditional consolidation engines. Copyright © 2017 John Wiley & Sons, Ltd.     

Precise contention-aware performance prediction on virtualized multicore system

Multicore systems are widely deployed in both the embedded and the high end computing infrastructures. However, traditional virtualization systems can not effectively isolate shared micro architectural resources among virtual machines (VMs) running on multicore systems. CPU and memory intensive VMs contending for these resources will lead to serious performance interference, which makes virtualization systems less efficient and VM performance less stable. In this paper, we propose a contention-aware performance prediction model on the virtualized multicore systems to quantify the performance degradation of VMs. First, we identify the performance interference factors and design synthetic micro-benchmarks to obtain VM’s contention sensitivity and intensity features that are correlated with VM performance degradation. Second, based on the contention features, we build VM performance prediction model using machine learning techniques to quantify the precise levels of performance degradation. The proposed model can be used to optimize VM performance on multicore systems. Our experimental results show that the performance prediction model achieves high accuracy and the mean absolute error is 2.83%.     

Effects of dynamic isolation for full virtualized RTOS and GPOS guests

Industrial systems currently include not only control processing with real-time operating system (RTOS) but also information processing with general-purpose operating system (GPOS). Multicore-based virtualization is an attractive option to provide consolidated environment when GPOS and RTOS are put in service on a single hardware platform. Researches on this technology have predominantly focused on the schedulability of RTOS virtual machines (VMs) by completely dedicated physical-CPUs (pCPUs) but have rarely considered parallelism or the throughput of the GPOS. However, it is also important that the multicore-based hypervisor adaptively selects pCPU assignment policy to efficiently manage resources in modern industrial systems. In this paper, we report our study on the effects of dynamic isolation when two mixed criticality systems are working on one platform. Based on our investigation of mutual interferences between RTOS VMs and GPOS VMs, we found explicit effects of dynamic isolation by special events. While maintaining low RTOS VMs scheduling latency, a hypervisor should manage pCPUs assignment by event-driven and threshold-based strategies to improve the throughput of GPOS VMs. Furthermore, we deal with implicit negative effects of dynamic isolation caused by the synchronization inside a GPOS VM, then propose a process of urgent boosting with dynamic isolation. All our methods are implemented in a real hypervisor, KVM. In experimental evaluation with benchmarks and an automotive digital cluster application, we analyzed that proposed dynamic isolation guarantees soft real-time operations for RTOS tasks while improving the throughput of GPOS tasks on a virtualized multicore system.     

Virtual machine power measuring technique with bounded error in cloud environments

In virtualized datacenters, accurately measuring the power consumption of virtual machines (VMs) is the prerequisite to achieve the goal of fine-grained power management. However, existing VM power models can only provide power measurements with empirical accuracy and unbounded error. In this paper, we firstly formulize the co-relation between utilization and accuracy of power model, and compare two classes of VM power models; then we propose a novel VM power model which is based on a conception named relative performance monitoring counter (PMC); finally, based on the relative PMC power model, we propose a novel VM scheduling algorithm which uses the information of relative PMC to compensate the recursive power consumption. Theoretical analysis indicates that the proposed algorithm can provide bounded error when measuring per-VM power consumption. Extensive experiments are conducted by using various benchmarks on different platforms, and the results show that the error of per-VM power measurement can be significantly reduced. In addition, the proposed algorithm is effective to improve the power efficiency of a server when its virtualization ratio is high.     

可扩展并行Web Server集群技术

采用并行Ｗｅｂ Ｓｅｒｖｅｒ集群技术实现高性能Ｗｅｂ Ｓｅｒｖｅｒ已经成为一种趋势。该技术具有性能高、可扩展性好、可靠性高、成本低等优点。本文介绍了我们研制的ＴＨ－Ｗｅｂ Ｃｌｕｓｔｅｒ的工作原理、组成结构和所采用的几种关键技术,并与现有的技术和方法进行了比较。本文还简要介绍了ＴＨ－Ｗｅｂ Ｃｌｕｓｔｅｒ上开发的两个应用系统ＴＨ－Ｗｅｂ Ｄｉｇｇｅｒ（信息挖掘工具）和ＴＨ－Ｗｅｂ Ｓｅａｒｃｈ（搜     

虚拟环境下Web服务动态负载均衡策略改进

为了提高Web服务集群的伸缩性和自动化能力,从虚拟化和负载均衡两方面研究集群系统,对现有负载采集策略做了改进,设计并实现了一种可根据负载值自动控制集群规模的模型XCluster。新模型运行在Xen提供的虚拟化环境中,实时监视宿主机层和虚拟机层的负载状态,随着集群系统总负载的增长,逐渐引入新的虚拟机来扩大集群规模,同时将任务合理分配到各个虚拟机节点上;当总负载下降时,逐渐关闭虚拟机缩小集群规模,释放出来的硬件资源又可以提供给其他集群系统使用。理论分析和实验结果表明,XCluster只需占用很少的网络通信量完成信息收集和命令下达,能够充分利用虚拟机易于管理的优势完成后端节点的调度,并且在任务总量相同的情况下,使用尽可能少的集群节点来执行任务。     

Efficient consolidation-aware VCPU scheduling on multicore virtualization platform

Multicore processors are widely used in today’s computer systems. Multicore virtualization technology provides an elastic solution to more efficiently utilize the multicore system. However, the Lock Holder Preemption (LHP) problem in the virtualized multicore systems causes significant CPU cycles wastes, which hurt virtual machine (VM) performance and reduces response latency. The system consolidates more VMs, the LHP problem becomes worse. In this paper, we propose an efficient consolidation-aware vCPU (CVS) scheduling scheme on multicore virtualization platform. Based on vCPU over-commitment rate, the CVS scheduling scheme adaptively selects one algorithm among three vCPU scheduling algorithms: co-scheduling, yield-to-head, and yield-to-tail based on the vCPU over-commitment rate because the actions of vCPU scheduling are split into many single steps such as scheduling vCPUs simultaneously or inserting one vCPU into the run-queue from the head or tail. The CVS scheme can effectively improve VM performance in the low, middle, and high VM consolidation scenarios. Using real-life parallel benchmarks, our experimental results show that the proposed CVS scheme improves the overall system performance while the optimization overhead remains low.     

LESSON: A system for lecture notes searching and sharing over Internet

In this paper, we present a system LESSON for lecture notes searching and sharing, which is dedicated to both instructors and students for effectively supporting their Web-based teaching and learning activities. The LESSON system employs a metasearch engine for lecture notes searching from Web and a peer-to-peer (P2P) overlay network for lecture notes sharing among the users. A metasearch engine provides an unified access to multiple existing component search engines and has better performance than general-purpose search engines. With the help of a P2P overlay network, all computers used by instructors and students can be connected into a virtual society over the Internet and communicate directly with each other for lecture notes sharing, without any centralized server and manipulation. In order to merge results from multiple component search engines into a single ranked list, we design the RSF strategy that takes rank, similarity and features of lecture notes into account. To implement query routing decision for effectively supporting lecture notes sharing, we propose a novel query routing mechanism. Experimental results indicate that the LESSON system has better performance in lecture notes searching from Web than some popular general-purpose search engines and some existing metasearch schemes; while processing queries within the system, it outperforms some typical routing methods. Concretely, it can achieve relatively high query hit rate with low bandwidth consumption in different types of network topologies.     

Developing resource consolidation frameworks for moldable virtual machines in clouds

This paper considers the scenario where multiple clusters of Virtual Machines (i.e., termed Virtual Clusters) are hosted in a Cloud system consisting of a cluster of physical nodes. Multiple Virtual Clusters (VCs) cohabit in the physical cluster, with each VC offering a particular type of service for the incoming requests. In this context, VM consolidation, which strives to use a minimal number of nodes to accommodate all VMs in the system, plays an important role in saving resource consumption. Most existing consolidation methods proposed in the literature regard VMs as “rigid” during consolidation, i.e., VMs’ resource capacities remain unchanged. In VC environments, QoS is usually delivered by a VC as a single entity. Therefore, there is no reason why VMs’ resource capacity cannot be adjusted as long as the whole VC is still able to maintain the desired QoS. Treating VMs as “moldable” during consolidation may be able to further consolidate VMs into an even fewer number of nodes. This paper investigates this issue and develops a Genetic Algorithm (GA) to consolidate moldable VMs. The GA is able to evolve an optimized system state, which represents the VM-to-node mapping and the resource capacity allocated to each VM. After the new system state is calculated by the GA, the Cloud will transit from the current system state to the new one. The transition time represents overhead and should be minimized. In this paper, a cost model is formalized to capture the transition overhead, and a reconfiguration algorithm is developed to transit the Cloud to the optimized system state with low transition overhead. Experiments have been conducted to evaluate the performance of the GA and the reconfiguration algorithm.     

Flubber: Two-level disk scheduling in virtualized environment

While virtualization enables multiple virtual machines (VMs)—with multiple operating systems and applications—to run within a physical server, it also complicates resource allocations trying to guarantee Quality of Service (QoS) requirements of the diverse applications running within these VMs. As QoS is crucial in the cloud, considerable research efforts have been directed towards CPU, memory and network allocations to provide effective QoS to VMs, but little attention has been devoted to disk resource allocation.This paper presents the design and implementation of Flubber, a two-level scheduling framework that decouples throughput and latency allocation to provide QoS guarantees to VMs while maintaining high disk utilization. The high-level throughput control regulates the pending requests from the VMs with an adaptive credit-rate controller, in order to meet the throughput requirements of different VMs and ensure performance isolation. Meanwhile, the low-level latency control, by the virtue of the batch and delay earliest deadline first mechanism (BD-EDF), re-orders all pending requests from VMs based on their deadlines, and batches them to disk devices taking into account the locality of accesses across VMs. We have implemented Flubber and made extensive evaluations on a Xen-based host. The results show that Flubber can simultaneously meet the different service requirements of VMs while improving the efficiency of the physical disk. The results also show improvement of up to 25% in the VM performance over state-of-art approaches: for example, in contract to the default Xen disk I/O scheduler—Completely Fair Queueing (CFQ)—besides achieving the desired QoS of each VM, Flubber speeds up the sequential and random reads by 17% and 25%, respectively, due to the efficient physical disk utilization.     

An Optimization-Based Scheme for Efficient Virtual Machine Placement

According to the important methodology of convex optimization theory, the energy-efficient and scalability problems of modern data centers are studied. Then a novel virtual machine (VM) placement scheme is proposed for solving these problems in large scale. Firstly, by referring the definition of VM placement fairness and utility function, the basic algorithm of VM placement which fulfills server constraints of physical machines is discussed. Then, we abstract the VM placement as an optimization problem which considers the inherent dependencies and traffic between VMs. By given the structural differences of recently proposed data center architectures, we further investigate a comparative analysis on the impact of the network architectures, server constraints and application dependencies on the potential performance gain of optimization-based VM placement. Comparing with the existing schemes, the performance improvements are illustrated from multiple perspectives, such as reducing the number of physical machines deployment, decreasing communication cost between VMs, improving energy-efficient and scalability of data centers.     

Denial-of-Service attacks on PCI passthrough devices: Demonstrating the impact on network- and storage-I/O performance

PCI Passthrough is an established x86 server technology for directly assigning PCIe devices to Virtual Machines (VMs). In combination with Single Root I/O Virtualization, which enables concurrent sharing of single physical PCIe I/O devices, PCI Passthrough enables low overhead and high performance I/O virtualization. Besides server environments, the combination is also a promising approach for sharing I/O in future multi-core embedded systems. In this paper, we demonstrate that PCI Passthrough has yet-to-be-solved problems regarding performance isolation, because it is prone to Denial-of-Service (DoS) attacks. VMs executing DoS attacks on Passthrough devices can degrade the I/O performance of devices that share PCIe links with the DoS victim, which may affect concurrent VMs and the host. We evaluate how attacks on an SR-IOV capable Gigabit Ethernet NIC cause a degradation of the system’s network- and storage-I/O performance. The attacked NIC’s TCP throughput drops by 35%; other NICs that share PCIe links with the victim see degradations of 46% and 65%; performance of a host-assigned SSD degrades by 77%. We investigate what influences the severity of such attacks and introduce three protection approaches.     

Adapting Web pages for small-screen devices

We propose a page-adaptation technique that splits existing Web pages into smaller, logically related units. To do this, we must first solve two technical problems: how to detect an existing Web page's semantic structure, and how to split a Web page into smaller blocks based on that structure. To date, we've implemented our technique in Web browsers for mobile devices, in a proxy server for adapting Web pages on the fly, and as an authoring tool plug-in for converting existing Web pages. The Web page can then be adapted to form a two-level hierarchy with a thumbnail representation at the top level for providing a global view and an index to a set of subpages at the bottom level for detailed information.     

Power-efficient assignment of virtual machines to physical machines

Motivated by current trends in cloud computing, we study a version of the generalized assignment problem where a set of virtual processors has to be implemented by a set of identical processors. For literature consistency, we say that a set of virtual machines (VMs) is assigned to a set of physical machines (PMs). The optimization criterion is to minimize the power consumed by all the PMs. We term the problem Virtual Machine Assignment (VMA). Crucial differences with previous work include a variable number of PMs, that each VM must be assigned to exactly one PM (i.e., VMs cannot be implemented fractionally), and a minimum power consumption for each active PM. Such infrastructure may be strictly constrained in the number of PMs or in the PMs’ capacity, depending on how costly (in terms of power consumption) it is to add a new PM to the system or to heavily load some of the existing PMs. Low usage or ample budget yields models where PM capacity and/or the number of PMs may be assumed unbounded for all practical purposes. We study four VMA problems depending on whether the capacity or the number of PMs is bounded or not. Specifically, we study hardness and online competitiveness for a variety of cases. To the best of our knowledge, this is the first comprehensive study of the VMA problem for this cost function.     

Online slack consolidation in global-EDF for energy consumption minimisation

Leakage power dissipation is one of the major concerns in homogeneous multicore platforms. Therefore, individual cores on such platforms are often equipped with multiple sleep states to reduce the leakage power dissipation. With the current body of knowledge, an efficient selection of sleep states is a non-trivial problem for system designers. In this work, we propose leakage-aware energy management algorithms for homogeneous multicore platforms using a global-EDF scheduler. Global-EDF assumes that at any time instant the tasks (constituting the application) with the closest absolute deadlines are selected for execution on any core of the platform, sometimes allowing migration. Initially, individual cores are allowed to change their power states independently. This assumption is relaxed in the second algorithm and cores transition into different power states in coordination with each other. The main idea behind the proposed algorithms consists of exploiting the spare capacity available in the schedule of each core to either initiate a sleep state on this core or prolong the sleep state of cores already in a sleep state in order to minimise the leakage power dissipation. The presented algorithms have low complexity, thus making it practically feasible. Evaluations are carried out by assuming the specifications of Intel Xeon E3-1285L V4 embedded multicore processor and Freescale P5040 QorIQ Integrated Processor to demonstrate its effectiveness. In the best-case, up to 50% and 60% of the energy consumption wasted in idle intervals — i.e., when a core is not performing any execution — on Intel Xeon and Freescale P5040 platform, respectively, is saved over the baseline global-EDF schedule.     

Efficient data distribution in a Web server farm

High-performance Web sites rely on Web server `farms', hundreds of computers serving the same content, for scalability, reliability, and low-latency access to Internet content. Deploying these scalable farms typically requires the power of distributed or clustered file systems. Building Web server farms on file systems complements hierarchical proxy caching. Proxy caching replicates Web content throughout the Internet, thereby reducing latency from network delays and off-loading traffic from the primary servers. Web server farms scale resources at a single site, reducing latency from queuing delays. Both technologies are essential when building a high-performance infrastructure for content delivery. The authors present a cache consistency model and locking protocol customized for file systems that are used as scalable infrastructure for Web server farms. The protocol takes advantage of the Web's relaxed consistency semantics to reduce latencies and network overhead. Our hybrid approach preserves strong consistency for concurrent write sharing with time-based consistency and push caching for readers (Web servers). Using simulation, we compare our approach to the Andrew file system and the sequential consistency file system protocols we propose to replace     

Optimal configuration of a multicore server processor for managing the power and performance tradeoff

We consider the problem of power and performance management for a multicore server processor in a cloud computing environment by optimal server configuration for a specific application environment. The motivation of the study is that such optimal virtual server configuration is important for dynamic resource provision in a cloud computing environment to optimize the power and performance tradeoff for certain specific type of applications. Our strategy is to treat a multicore server processor as an M/M/m queueing system with multiple servers. The system performance measures are the average task response time and the average power consumption. Two core speed and power consumption models are considered, namely, the idle-speed model and the constant-speed model. Our investigation includes justification of centralized management of computing resources, server speed constrained optimization, power constrained performance optimization, and performance constrained power optimization. Our main results are (1) cores should be managed in a centralized way to provide the highest performance without consumption of more energy in cloud computing; (2) for a given server speed constraint, fewer high-speed cores perform better than more low-speed cores; furthermore, there is an optimal selection of server size and core speed which can be obtained analytically, such that a multicore server processor consumes the minimum power; (3) for a given power consumption constraint, there is an optimal selection of server size and core speed which can be obtained numerically, such that the best performance can be achieved, i.e., the average task response time is minimized; (4) for a given task response time constraint, there is an optimal selection of server size and core speed which can be obtained numerically, such that minimum power consumption can be achieved while the given performance guarantee is maintained.     

Power-efficient distributed scheduling of virtual machines using workload-aware consolidation techniques

There is growing demand on datacenters to serve more clients with reasonable response times, demanding more hardware resources, and higher energy consumption. Energy-aware datacenters have thus been amongst the forerunners to deploy virtualization technology to multiplex their physical machines (PMs) to as many virtual machines (VMs) as possible in order to utilize their hardware resources more effectively and save power. The achievement of this objective strongly depends on how smart VMs are consolidated. In this paper, we show that blind consolidation of VMs not only does not reduce the power consumption of datacenters but it can lead to energy wastage. We present four models, namely the target system model, the application model, the energy model, and the migration model, to identify the performance interferences between processor and disk utilizations and the costs of migrating VMs. We also present a consolidation fitness metric to evaluate the merit of consolidating a number of known VMs on a PM based on the processing and storage workloads of VMs. We then propose an energy-aware scheduling algorithm using a set of objective functions in terms of this consolidation fitness metric and presented power and migration models. The proposed scheduling algorithm assigns a set of VMs to a set of PMs in a way to minimize the total power consumption of PMs in the whole datacenter. Empirical results show nearly 24.9% power savings and nearly 1.2% performance degradation when the proposed scheduling algorithm is used compared to when other scheduling algorithms are used.     

普通高考招生信息网站的负载均衡

高考信息的网上收集与发布,存在访问规模大并且时间分布严重不均,导致Web服务器不堪重负的问题。基于LVS项目,给出了采用DR模式的Web集群的全面解决方案,并应用于2005年海南省普通高招报考系统,收到良好效果。最后以实验测试并证明了Web集群的优异性能。