Kejiang Ye

Virtual Machine Based Energy-Efficient Data Center Architecture for Cloud Computing: A Performance Perspective

Virtual machine technology is widely applied to modern data center for cloud computing as a key technology to realize energy-efficient operation of servers. Server consolidation achieves energy efficiency by enabling multiple instantiations of operating systems (OSes) to run simultaneously on a single physical machine. While, live migration of virtual machine can transfer the virtual machine workload from one physical machine to another without interrupting service. However, both the two technologies have their own performance overheads. There is a tradeoff between the performance and energy efficiency. In this paper, we study the energy efficiency from the performance perspective. Firstly, we present a virtual machine based energy-efficient data center architecture for cloud computing. Then we investigate the potential performance overheads caused by server consolidation and live migration of virtual machine technology. Experimental results show that both the two technologies can effectively implement energy-saving goals with little performance overheads. Efficient consolidation and migration strategies can improve the energy efficiency.

Analyzing and Modeling the Performance in Xen-Based Virtual Cluster Environment

Virtualization technology is currently widely used due to its benefits on high resource utilization, flexible manageability and powerful system security. However, its use for high performance computing (HPC) is still not popular due to the unclearness of the virtualization overheads. Its worthy to evaluate the virtualization cost and to find the performance bottleneck when running HPC applications in virtual cluster. We first evaluate the basic performance overheads due to virtualization. Then we create a 16-node virtual cluster and perform a performance evaluation for both para-virtualization and full virtualization. After that, we evaluate the MPI (Message Passing Interface) scalability to investigate the impact of MPI and network communication between virtual machines. In addition to the macro assessment, we use the Oprofile/Xenoprof to investigate the architecture characterization like CPU cycle, L2 cache misses, DTLB misses and ITLB misses which is an auxiliary explanation to the performance bottleneck. Experimental results indicate that performance overheads of virtualization are acceptable for HPC, para-virtualization is very suitable for HPC due to the high virtualization efficiency and efficient inter-domain communication. Finally, we use the non-linear regression modeling technology to present a performance model for network latency and bandwidth to predict the performance in virtual cluster environment.

Profiling-Based Workload Consolidation and Migration in Virtualized Data Centers

Improving energy efficiency of data centers has become increasingly important nowadays due to the significant amounts of power needed to operate these centers. An important method for achieving energy efficiency is server consolidation supported by virtualization. However, server consolidation may incur significant degradation to workload performance due to virtual machine (VM) co-location and migration. How to reduce such performance degradation becomes a critical issue to address. In this paper, we propose a profiling-based server consolidation framework which minimizes the number of physical machines (PMs) used in data centers while maintaining satisfactory performance of various workloads. Inside this framework, we first profile the performance losses of various workloads under two situations: running in co-location and experiencing migrations. We then design two modules: (1) consolidation planning module which, given a set of workloads, minimizes the number of PMs by an integer programming model, and (2) migration planning module which, given a source VM placement scenario and a target VM placement scenario, minimizes the number of VM migrations by a polynomial time algorithm. Also, based on the workload performance profiles, both modules can guarantee the performance losses of various workloads below configurable thresholds. Our experiments for workload profiling are conducted with real data center workloads and our experiments on our two modules validate the integer programming model and the polynomial time algorithm.

Virt-LM: a benchmark for live migration of virtual machine

Virtualization technology has been widely applied in data centers and IT infrastructures, with advantages of server consolidation and live migration. Through live migration, data centers could flexibly move virtual machines among different physical machines to balance workloads, reduce energy consumption and enhance service availability. Todays data centers can grow to a huge scale. This implies that frequent live migration would be desirable for the economic use of hardware resources. Then, the performance of the live migration strategy will be an issue. So, we need a reliant evaluation method to choose the software and hardware environments that will produce the best live migration performance. However, there is not a complete live migration benchmark available currently. In addition, the existing evaluation methodologies select different metrics, different workloads and different test means. Thus, it is difficult to compare their results. In this paper we first survey the current research and their evaluation methods on live migration. We then summarize the critical issues for the live migration evaluation and also raise other unreported potential problems. We propose our solutions and present an implementation in our live migration benchmark -- Virt-LM. This is a benchmark for comparing live migration performance among different software and hardware environments in a data center scenario. We detail its design and provide some experimental results to validate its effectiveness.

vHadoop: A Scalable Hadoop Virtual Cluster Platform for MapReduce-Based Parallel Machine Learning with Performance Consideration

Big data processing is currently becoming increasingly important in modern era due to the continuous growth of the amount of data generated by various fields such as particle physics, human genomics, earth observation, etc. However, the efficiency of processing large-scale data on modern virtual infrastructure, especially on the virtualized cloud computing infrastructure, is not clear. This paper focuses on the performance of hadoop virtual cluster and proposes a scalable hadoop virtual cluster platform vHadoop for the large-scale MapReduce-based parallel data processing. We first describe the design and implementation of vHadoop platform. Then we perform a series of experiments to investigate both the static and dynamic performance of vHadoop platform, such as the performance characterization of cross-domain hadoop virtual cluster and live migraiton of hadoop virtual cluster. After that, we use the vHadoop platform to process 6 typical parallel clustering algorithms, such as Canopy, Dirichlet, Fuzzy k-Means, k-Means, Mean Shift, MinHash, etc, on two typical datasets. Experimental results verify the efficiency of vHadoop platform to process the MapReduce-based parallel machine learning applications.

Two Optimization Mechanisms to Improve the Isolation Property of Server Consolidation in Virtualized Multi-core Server

Virtualization brings many benefits such as improving system utilization and reducing cost through server consolidation. However, it also introduces isolation problem when running multiple virtual machine workloads in one physical platform. Additionally, with the advent of multi-core technology, more and more cores are built into one die in todays data center that will share and compete for the resource like cache. Its worthy to study the isolation of server consolidation in modern multi-core platform. However, to our knowledge there are few work done on the isolation property especially the fault isolation property when one of the virtual machine workloads is attacked in server consolidation. In this paper, we study the isolation property from performance perspective and provide two optimization methods to improve the isolation property. We first define the isolation property and quantify the performance isolation in consolidation and propose a VM-level optimization method. Then we study the fault isolation by introducing a misbehavior virtual machine in server consolidation scenario and propose a core-level cache-aware optimization method to improve the fault isolation. Experimental results show that our two optimization methods can effectively improve the performance isolation and fault isolation with 29.39% and 19.52% respectively. Whats more, Oprofile/Xenoprof toolkits are used to find out the factors affecting isolation property from the hardware events level.

DartCSim+: Enhanced CloudSim with the Power and Network Models Integrated

CloudSim is one of the most powerful simulation platforms for cloud computing. It supports the energy-conscious scheduling and network simulation in the latest version. However, it still faces several limitations: 1) Current CloudSim cannot support both the power model and the network model at the same time. 2) The network components in current CloudSim do not support power-aware simulation. 3) The simulation of migration does not take into account the network overheads. To overcome these limitations, we design and implement an enhanced cloud simulation platform called DartCSim+ that supports the energy-aware network simulation and network-aware live migration. Further, we also implement a resubmit mechanism for packets transmission to provide a more real network behavior to solve transmission failure which is caused by migration or network failure. Finally, three groups of experiments are performed to demonstrate the effectiveness of DartCSim+.

vTestkit: A Performance Benchmarking Framework for Virtualization Environments

Virtualization technology has attracted wide attention in recent years as a method to improve resource utilization, reduce costs, and ease server management. However, the performance penalty resulting from virtualization is an unneglectable problem and should be carefully evaluated. To our knowledge, there are few performance evaluating tools developed for virtualization environments. We propose a configurable framework and implement a prototype vTestkit to provide a platform to do performance evaluation for virtualization environments easily, flexibly, and automatically. In this paper, we first discuss the requirements and challenges of performance measurement in virtualization environments, and then present a methodology for characterizing the performance of single virtual machine (VM) scenario and multi-VM scenario. Then we introduce the architecture of vTestkit framework, implement details, and the testing process with vTestkit. Finally, three typical case studies are presented to show that vTestkit can meet the complex testing requirements well and is propitious to various scenarios.

Informed live migration strategies of virtual machines for cluster load balancing

Virtualization technology brings great conveniences to cluster and data center management. By using this technique, we can reconstruct a new computing environment quickly and easily. Compared to the traditional cluster environment, load balancing in a virtualized environment needs to address several new problems. This paper focuses on live migration strategies for load balancing in the virtualized cluster. We first divide the whole balancing process into three sub-problems, namely, the selection of the VM being migrated to, the choice of destination host and the determination of the migration execution sequence. Then we perform a series of experiments to investigate the particular features of the live migration of virtual machines in the balancing scenario. Based on our experiment results, we propose an informed live migration strategy which includes affinity-aware decision making and workload-aware migration to improve the efficiency of configuration of the virtualized cluster.

Performance combinative evaluation of typical virtual machine monitors

As one of pivotal components, the efficiency of virtual machine monitor(VMM) will largely impact the performance of a virtualization system. Therefore, evaluating the performance of VMMs adopting different virtualization technologies becomes more and more important. This paper selects three typical open source VMMs(i.e. OpenVZ, Xen and KVM) as the delegates of operating system-level virtualization, para-virtualization and full-virtualization to evaluate their performance with a combinative method that measures their macro-performance and micro-performance as a black box and analyzes their performance characteristic as a white box. By correlating the analysis results of two-granularity performance data, some potential performance bottlenecks come out.