MapReduce集群环境下的数据放置策略

MapReduce是一种适用于大规模数据密集型应用的有效编程模型,具有编程简单、易于扩展、容错性好等特点,已在并行和分布式计算领域得到了广泛且成功的应用.由于MapReduce将计算扩展到大规模的机器集群上,处理数据的合理放置成为影响MapReduce集群系统性能(包括能耗、资源利用率、通信和I/O代价、响应时间、系统的可靠性和吞吐率等)的关键因素之一.首先,对MapReduce编程模型的典型实现——Hadoop缺省的数据放置策略进行分析,并进一步讨论了MapReduce框架下,设计数据放置策略时需考虑的关键问题和衡量数据放置策略的标准;其次,对目前MapReduce集群环境下的数据放置策略优化方法的研究与进展进行了综述和分析;最后,分析和归纳了MapReduce集群环境下数据放置策略的下一步研究工作.     

Tuning small analytics on Big Data: Data partitioning and secondary indexes in the Hadoop ecosystem

In the recent years the problems of using generic storage (i.e., relational) techniques for very specific applications have been detected and outlined and, as a consequence, some alternatives to Relational DBMSs (e.g., HBase) have bloomed. Most of these alternatives sit on the cloud and benefit from cloud computing, which is nowadays a reality that helps us to save money by eliminating the hardware as well as software fixed costs and just pay per use. On top of this, specific querying frameworks to exploit the brute force in the cloud (e.g., MapReduce) have also been devised. The question arising next tries to clear out if this (rather naive) exploitation of the cloud is an alternative to tuning DBMSs or it still makes sense to consider other options when retrieving data from these settings.In this paper, we study the feasibility of solving OLAP queries with Hadoop (the Apache project implementing MapReduce) while benefiting from secondary indexes and partitioning in HBase. Our main contribution is the comparison of different access plans and the definition of criteria (i.e., cost estimation) to choose among them in terms of consumed resources (namely CPU, bandwidth and I/O).     

基于MapReduce的量子蚁群算法

量子蚁群算法是在蚁群算法的基础上结合量子计算而提出的,该算法具有较好的全局寻优能力和种群多样性。应用MapReduce的key/value编程模型,将量子蚁群算法并行化,提出了基于MapReduce的量子蚁群算法（MQACA）,并将其部署到Hadoop云计算平台上运行。对0-1背包问题的测试结果证明,随着数据规模的扩大和并行程度的提高,MQACA具有良好的加速比和并行效率。     

Governing energy consumption in Hadoop through CPU frequency scaling: An analysis

With increasingly inexpensive storage and growing processing power, the cloud has rapidly become the environment of choice to store and analyze data for a variety of applications. Most large-scale data computations in the cloud heavily rely on the MapReduce paradigm and on its Hadoop implementation. Nevertheless, this exponential growth in popularity has significantly impacted power consumption in cloud infrastructures. In this paper, we focus on MapReduce processing and we investigate the impact of dynamically scaling the frequency of compute nodes on the performance and energy consumption of a Hadoop cluster. To this end, a series of experiments are conducted to explore the implications of Dynamic Voltage and Frequency Scaling (DVFS) settings on power consumption in Hadoop clusters. By enabling various existing DVFS governors (i.e., performance, powersave, ondemand, conservative and userspace) in a Hadoop cluster, we observe significant variation in performance and power consumption across different applications: the different DVFS settings are only sub-optimal for several representative MapReduce applications. Furthermore, our results reveal that the current CPU governors do not exactly reflect their design goal and may even become ineffective to manage the power consumption in Hadoop clusters. This study aims at providing a clearer understanding of the interplay between performance and power management in Hadoop clusters and therefore offers useful insight into designing power-aware techniques for Hadoop systems.     

A novel real-time scheduling algorithm and performance analysis of a MapReduce-based cloud

MapReduce, a popular programming model for processing data-intensive tasks, has achieved great success in a wide range of applications such as search indexing, social network mining, collaborative recommendation, and spam detection. However, the ability of MapReduce is limited in two respects by its default schedulers. First, it does not support concurrent services sharing a cloud datacenter and second, it fails to guarantee response time for deadline-constrained services. This paper proposes the Paused Rate Monotonic (PRM) algorithm for scheduling hard real-time tasks on a MapReduce-based cloud. The scheduling performance is analyzed theoretically. We prove a bound on cluster utilization, which can be used as a sufficient condition to test whether a given task set can be scheduled. Both the theoretical analysis and experimental evaluation show that the PRM algorithm outperforms traditional real-time ones by improving the probability that a real-time task set can be scheduled on a MapReduce-based cloud.     

基于云计算异构环境的Hadoop性能分析

针对当前云服务发展中遇到的数据量急速增加为后台数据中心带来的数据处理问题,本文实施了Hadoop 的 性能测试。首先对云计算异构环境的Hadoop 性能进行了描述,其次对Hadoop 集群中两种节点的性能差别进行了分析,并给 出了实验异构的平台的设计和实验结果。     

Towards a framework for large-scale multimedia data storage and processing on Hadoop platform

Cloud computing techniques take the form of distributed computing by utilizing multiple computers to execute computing simultaneously on the service side. To process the increasing quantity of multimedia data, numerous large-scale multimedia data storage computing techniques in the cloud computing have been developed. Of all the techniques, Hadoop plays a key role in the cloud computing. Hadoop, a computing cluster formed by low-priced hardware, can conduct the parallel computing of petabytes of multimedia data. Hadoop features high-reliability, high-efficiency, and high-scalability. The numerous large-scale multimedia data computing techniques include not only the key core techniques, Hadoop and MapReduce, but also the data collection techniques, such as File Transfer Protocol and Flume. In addition, distributed system configuration allocation, automatic installation, and monitoring platform building and management techniques are all included. As a result, only with the integration of all the techniques, a reliable large-scale multimedia data platform can be offered. In this paper, we introduce how cloud computing can make a breakthrough by proposing a multimedia social network dataset on Hadoop platform and implementing a prototype version. Detailed specifications and design issues are discussed as well. An important finding of this article is that we can save more time if we conduct the multimedia social networking analysis using Cloud Hadoop Platform rather than using a single computer. The advantages of cloud computing over the traditional data processing practices are fully demonstrated in this article. The applicable framework designs and the tools available for the large-scale data processing are also proposed. We show the experimental multimedia data including data sizes and processing time.     

Hadoop平台下自响应故障感知的检测技术研究实现

当今云计算环境下,Hadoop已经成为大数据处理的事实标准。然而云计算具有大规模、高复杂和动态性的特点,容易导致故障的发生,影响Hadoop上运行的作业。虽然Hadoop具有内置的故障检测和恢复机制,但云环境中不同节点负载大小的变化,被调度的作业仍然导致失败。针对此问题提出自响应故障感知的检测调度方法,对异构环境负载能力的不同,而做出服务器快节点和慢节点的判断,把作业分配调度到合适的节点上执行,调整任务决策来尽可能的防止任务失败的发生。最后在Hadoop框架下与基本调度器进行实验性能比较,结果显示该方法减少作业失败率最高达19%,并缩短了作业执行时间,同时也减少CPU和内存的使用。     

不同MapReduce运行系统的性能测试与分析

随着云计算技术的发展,许多MapReduce运行系统被开发出来,如Hadoop、Phoenix和Twister等.直观上,Hadoop具有很强的可扩展性、稳定性,适合处理大规模离线应用;Phoenix具有运行速度快等优点,适合处理数据密集型任务;Twister是轻量级的迭代系统,非常适合迭代式的应用.不同的应用在不同的MapReduce运行系统中有着不同的性能.通过测试不同应用在这些运行系统上的性能,给出了实验比较和性能分析,从而为大数据处理时选择合适的并行编程模型提供依据.     

Sports performance prediction model based on integrated learning algorithm and cloud computing Hadoop platform

This article discusses the classification and research performance information properties. It also discusses construction and application of the Hadoop cloud computing platform. The model presented in this article is a one piece learning algorithm which is a predictive model and a model of cloud based data collection. This model is supported by Hadoop which is suitable for computing with different data sizes. A large number of simulations are performed on the Hadoop platform, under different working conditions, to verify the accuracy and characteristics of the training skill. Spark framework of this research is to develop computational engine efficiency and improve rain prediction models successfully and effectively using big data and Hadoop learning. Therefore, the planned high timeliness and accuracy of real-time hurricane forecast with rain, can solve the problem.     

Hadoop云平台MapReduce模型优化研究

针对Hadoop平台MapReduce分布式计算模型运行机制中的顺序制约而产生的计算资源浪费问题,从提高平台中每个执行节点的细粒度并行数据处理角度出发,结合Java共享内存多线程编程技术,对该模型进行了优化,提出一种MapReduce+OpenMP粗细粒度相结合的分布式并行计算模型。并在由四个节点组成的Hadoop集群环境下对不同规模大小的出租车GPS轨迹数据分析处理,验证该模型的性能和效率,实验结果证明MapReduce+OpenMP分布式并行计算模型确实能够提高针对大数据集的计算效率,是对Hadoop平台大数据分析处理模型有效的完善和优化。     

基于最小生成树的大规模数据分类模型及其MapReduce实现

数据的快速增长,为我们提供了更多的信息,然而,也对传统信息获取技术提出了挑战。这篇论文提出了MCMM算法,它是基于MapReduce的大规模数据分类模型的最小生成树（MST）的算法。它可以看做是介于传统的KNN方法和基于聚类分类方法之间的模型,旨在克服这两种方法的不足并能处理大规模的数据。在这一模型中,训练集作为有权重的无向完全图来处理。顶点是对象,两点之间边的权重是对象间的距离。这一距离,不同于欧几里得距离,它是一个特定的距离度量。这样,可以找到图中最小生成树集,其中,图中每棵树代表一个类。为了降低时间复杂度,提取了每棵树中最具代表性的点来代表该树。这些压缩了的点集,可以通过计算无标签对象和它们之间的距离,来进行分类。MCMM模型基于MapReduce实现并且部署在Hadoop平台。该模型可扩展处理大规模的数据,是因为Hadoop支持数据密集分布应用,并且这些应用可以和数以千计的节点和数据一起运作。另外,MapReduce 和Hadoop能在由商品机组成的集群上很好的运行。MCMM模型使用云平台并且通过使用MapReduce 和Hadoop进行云计算是有益处的。实验采用的数据集包括从UCI数据库得到的真实数据和一些模拟数据,实验使用了4000个集群。实验表明,MCMM模型在精确度和扩展性上优于KNN和其他一些经常使用的基础分类方法。     

Constructing gazetteers from volunteered Big Geo-Data based on Hadoop

Traditional gazetteers are built and maintained by authoritative mapping agencies. In the age of Big Data, it is possible to construct gazetteers in a data-driven approach by mining rich volunteered geographic information (VGI) from the Web. In this research, we build a scalable distributed platform and a high-performance geoprocessing workflow based on the Hadoop ecosystem to harvest crowd-sourced gazetteer entries. Using experiments based on geotagged datasets in Flickr, we find that the MapReduce-based workflow running on the spatially enabled Hadoop cluster can reduce the processing time compared with traditional desktop-based operations by an order of magnitude. We demonstrate how to use such a novel spatial-computing infrastructure to facilitate gazetteer research. In addition, we introduce a provenance-based trust model for quality assurance. This work offers new insights on enriching future gazetteers with the use of Hadoop clusters, and makes contributions in connecting GIS to the cloud computing environment for the next frontier of Big Geo-Data analytics.     

基于MapReduce的并行同态加密算法

根据云计算分布式的特点,并结合同态加密和Hadoop环境下MapReduce并行框架,提出了一种基于MapReduce计算框架的并行同态加密方案。实现了具体的并行同态加密算法,并对该方案的安全性和正确性进行了理论分析。同时,在16个核的计算集群中进行实验,数据加密的加速比可以达到13。实验结果表明,基于MapReduce的同态加密方案可以有效地减少数据的加密时间,有利于面向实时的应用。     

基于Wi-Fi与Web的云计算资源调度算法研究

为缩短云计算执行时间,改善云计算性能,在一定程度上加强云计算资源节点完成任务成功率,需要对云计算资源进行调度;当前的云计算资源调度算法在进行调度时,通过选择合适的调度参数并利用CloudSim仿真工具,完成对云计算资源的调度;该算法在运行时无法有效地进行平衡负载,导致云计算资源调度的均衡性能较差,存在云计算资源调度结果误差大的问题;为此,提出一种基于Wi-Fi与Web的云计算资源调度算法;该算法首先利用自适应级联滤波算法对云计算资源数据流进行滤波降噪,然后以降噪结果为基础,采用本体论对云计算资源进行预处理操作,最后通过人工蜂群算法完成对云计算资源的调度;实验结果证明,所提算法可以良好地应用于云计算资源调度中,有效提高了云计算资源利用率,具有实用性以及可实践性,为该领域的后续研究发展提供了可靠支撑。     

Flame-MR: An event-driven architecture for MapReduce applications

Nowadays, many organizations analyze their data with the MapReduce paradigm, most of them using the popular Apache Hadoop framework. As the data size managed by MapReduce applications is steadily increasing, the need for improving the Hadoop performance also grows. Existing modifications of Hadoop (e.g., Mellanox Unstructured Data Accelerator) attempt to improve performance by changing some of its underlying subsystems. However, they are not always capable to cope with all its performance bottlenecks or they hinder its portability. Furthermore, new frameworks like Apache Spark or DataMPI can achieve good performance improvements, but they do not keep compatibility with existing MapReduce applications. This paper proposes Flame-MR, a new event-driven MapReduce architecture that increases Hadoop performance by avoiding memory copies and pipelining data movements, without modifying the source code of the applications. The performance evaluation on two representative systems (an HPC cluster and a public cloud platform) has shown experimental evidence of significant performance increases, reducing the execution time by up to 54% on the Amazon EC2 cloud.     

Performance modeling of big data applications in the cloud centers

Cloud computing has evolved as an efficient paradigm to process big data applications. Performance evaluation of cloud center is a necessary prerequisite to guarantee quality of service. However, it is a challenge task to effectively analyze the performance of cloud service due to the complexity of cloud resources and the diversity of big data applications. In this paper, we leverage queuing theory and probabilistic statistics to propose a performance evaluation model for cloud center under big data application arrivals. In this model, the tasks (i.e., big data applications) are with Poisson arrivals, each task is divided into lots of parallel subtasks, and the number of subtasks follows a general distribution. The model allows to calculate the important performance indicators such as mean number of subtasks in the system, the probability that a task obtains immediate service, task waiting time and blocking probability. The model can also be used to predict the time cost of performing application. Finally, we use the simulations and benchmarking running WordCount and TeraSort applications on a Hadoop platform to demonstrate the utility of the model.     

On construction of a distributed data storage system in cloud

In the past, people have focused on cluster computing and grid computing. Now, however, this focus has shifted to cloud computing. Irrespective of what techniques are used, there are always storage requirements. The challenge people face in this area is the huge amount of data to be stored, and its complexity. People are now using many cloud applications. As a result, service providers must serve increasingly more people, causing more and more connections involving substantially more data. These problems could have been solved in the past, but in the age of cloud computing, they have become more complex. This paper focuses on cloud computing infrastructure, and especially data services. The goal of this paper is to implement a high performance and load balancing, and able-to-be-replicated system that provides data storage for private cloud users through a virtualization system. This system extends and enhances the functionality of the Hadoop distributed system. The proposed approach also implements a resource monitor of machine status factors such as CPU, memory, and network usage to help optimize the virtualization system and data storage system. To prove and extend the usability of this design, a synchronize app was also developed running on Android based on our distributed data storage.     

用于Hadoop2.x的MapReduce性能评估模型

基于MapReduce的程序被越来越多地应用于大型数据分析的应用中.Apache Hadoop是最常用的开源MapReduce模型之一.程序运行时间的缩短对于MapReduce程序以及所有数据处理应用而言至关重要,而能够准确估算MapReduce程序的执行时间是优化程序的重要环节.本文定义了一个在Hadoop2.x版本...     

一种基于Hadoop的云运维监控模型设计与实现

随着企业信息化在生产实时监测、海量存储和科学分析决策等方面的需求不断提升,运维监控系统已逐渐成为主要的管理手段。采用最新的云计算技术,设计及搭建一个数据规模易扩展、处理速度快、安全性高、成本低的云运维监控系统;针对运维控制系统中海量监控历史数据实时提取响应速度慢的缺点,设计并实现一种基于Hadoop的分布式海量数据处理模型。仿真实验证明,Hadoop在对云监控系统中的海量数据提取效率优于传统方法,随着数据量的快速增长,优势越明显。