iMapReduce: A Distributed Computing Framework for Iterative Computation

Iterative computation is pervasive in many applications such as data mining, web ranking, graph analysis, online social network analysis, and so on. These iterative applications typically involve massive data sets containing millions or billions of data records. This poses demand of distributed computing frameworks for processing massive data sets on a cluster of machines. MapReduce is an example of such a framework. However, MapReduce lacks built-in support for iterative process that requires to parse data sets iteratively. Besides specifying MapReduce jobs, users have to write a driver program that submits a series of jobs and performs convergence testing at the client. This paper presents iMapReduce, a distributed framework that supports iterative processing. iMapReduce allows users to specify the iterative computation with the separated map and reduce functions, and provides the support of automatic iterative processing within a single job. More importantly, iMapReduce significantly improves the performance of iterative implementations by (1) reducing the overhead of creating new MapReduce jobs repeatedly, (2) eliminating the shuffling of static data, and (3) allowing asynchronous execution of map tasks. We implement an iMapReduce prototype based on Apache Hadoop, and show that iMapReduce can achieve up to 5 times speedup over Hadoop for implementing iterative algorithms.     

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.     

Functional Models of Hadoop MapReduce with Application to Scan

MapReduce, first proposed by Google, is a remarkable programming model for processing very large amounts of data. An open-source implementation of MapReduce, called Hadoop, is now used for developing a wide range of applications. Although developing a correct and efficient program on MapReduce is much easier than developing one with MPI etc., it is still nontrivial if the target application requires involved functionalities of Hadoop MapReduce. Under these situations, functional models for MapReduce computation play important roles because we can utilize them for better understanding, proving the correctness, and even optimization of MapReduce programs. In this paper, we develop two functional models, a low-level one and a high-level one, which capture the semantics of Hadoop MapReduce computation. We discuss the detailed semantics mainly in terms of the following two computations: the computation of Mapper and Reducer classes and the computation in the Shuffle phase with the secondary-sorting technique. In addition, we develop MapReduce algorithms for the scan computational pattern (prefix sums) on the newly proposed models.     

Adapting scientific computing problems to clouds using MapReduce

Cloud computing, with its promise of virtually infinite resources, seems to suit well in solving resource greedy scientific computing problems. To study this, we established a scientific computing cloud (SciCloud) project and environment on our internal clusters. The main goal of the project is to study the scope of establishing private clouds at the universities. With these clouds, students and researchers can efficiently use the already existing resources of university computer networks, in solving computationally intensive scientific, mathematical, and academic problems. However, to be able to run the scientific computing applications on the cloud infrastructure, the applications must be reduced to frameworks that can successfully exploit the cloud resources, like the MapReduce framework. This paper summarizes the challenges associated with reducing iterative algorithms to the MapReduce model. Algorithms used by scientific computing are divided into different classes by how they can be adapted to the MapReduce model; examples from each such class are reduced to the MapReduce model and their performance is measured and analyzed. The study mainly focuses on the Hadoop MapReduce framework but also compares it to an alternative MapReduce framework called Twister, which is specifically designed for iterative algorithms. The analysis shows that Hadoop MapReduce has significant trouble with iterative problems while it suits well for embarrassingly parallel problems, and that Twister can handle iterative problems much more efficiently. This work shows how to adapt algorithms from each class into the MapReduce model, what affects the efficiency and scalability of algorithms in each class and allows us to judge which framework is more efficient for each of them, by mapping the advantages and disadvantages of the two frameworks. This study is of significant importance for scientific computing as it often uses complex iterative methods to solve critical problems and adapting such methods to cloud computing frameworks is not a trivial task.     

SHadoop: Improving MapReduce performance by optimizing job execution mechanism in Hadoop clusters

As a widely-used parallel computing framework for big data processing today, the Hadoop MapReduce framework puts more emphasis on high-throughput of data than on low-latency of job execution. However, today more and more big data applications developed with MapReduce require quick response time. As a result, improving the performance of MapReduce jobs, especially for short jobs, is of great significance in practice and has attracted more and more attentions from both academia and industry. A lot of efforts have been made to improve the performance of Hadoop from job scheduling or job parameter optimization level. In this paper, we explore an approach to improve the performance of the Hadoop MapReduce framework by optimizing the job and task execution mechanism. First of all, by analyzing the job and task execution mechanism in MapReduce framework we reveal two critical limitations to job execution performance. Then we propose two major optimizations to the MapReduce job and task execution mechanisms: first, we optimize the setup and cleanup tasks of a MapReduce job to reduce the time cost during the initialization and termination stages of the job; second, instead of adopting the loose heartbeat-based communication mechanism to transmit all messages between the JobTracker and TaskTrackers, we introduce an instant messaging communication mechanism for accelerating performance-sensitive task scheduling and execution. Finally, we implement SHadoop, an optimized and fully compatible version of Hadoop that aims at shortening the execution time cost of MapReduce jobs, especially for short jobs. Experimental results show that compared to the standard Hadoop, SHadoop can achieve stable performance improvement by around 25% on average for comprehensive benchmarks without losing scalability and speedup. Our optimization work has passed a production-level test in Intel and has been integrated into the Intel Distributed Hadoop (IDH). To the best of our knowledge, this work is the first effort that explores on optimizing the execution mechanism inside map/reduce tasks of a job. The advantage is that it can complement job scheduling optimizations to further improve the job execution performance.     

Mapreduce performance model for Hadoop 2.x

MapReduce is a popular programming model for distributed processing of large data sets. Apache Hadoop is one of the most common open-source implementations of such paradigm. Performance analysis of concurrent job executions has been recognized as a challenging problem, at the same time, that may provide reasonably accurate job response time estimation at significantly lower cost than experimental evaluation of real setups. In this paper, we tackle the challenge of defining MapReduce performance model for Hadoop 2.x. While there are several efficient approaches for modeling the performance of MapReduce workloads in Hadoop 1.x, they could not be applied to Hadoop 2.x due to fundamental architectural changes and dynamic resource allocation in Hadoop 2.x. Thus, the proposed solution is based on an existing performance model for Hadoop 1.x, but taking into consideration architectural changes and capturing the execution flow of a MapReduce job by using queuing network model. This way, the cost model reflects the intra-job synchronization constraints that occur due the contention at shared resources. The accuracy of our solution is validated via comparison of our model estimates against measurements in a real Hadoop 2.x setup.     

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...     

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.     

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。     

云计算下的SPRINT并行算法研究

目前,由于云计算的出现,越来越多的中小企业在分析海量数据时能便利地找到廉价的解决方案。本文,鉴于MapReduce作为Hadoopd中的重要编程模型,在介绍基于云计算的Hadoop平台和数据挖掘技术中的SPRINT分类算法的基础上,详细描述SPRINT的并行算法在MapReduce编程模型上的执行流程,并利用研究出的决策树模型对输入数据进行分类。     

时空轨迹大数据分布式蜂群模式挖掘算法

针对时空轨迹大数据的蜂群模式挖掘需求,提出了一种高效的基于MapReduce的分布式蜂群模式挖掘算法。首先,提出了基于最大移动目标集的对象集闭合蜂群模式概念,并利用最小时间支集优化了串行挖掘算法;其次,提出了蜂群模式的并行化挖掘模型,利用蜂群模式时间域无关性,并行化了聚类与子时间域上的蜂群模式挖掘过程;第三,设计了一个基于MapReduce链式架构的分布式并行挖掘算法,通过四个阶段快速地实现了蜂群模式的并行挖掘;最后,在Hadoop平台上,使用真实交通轨迹大数据集对分布式算法的有效性和高效性进行了验证与分析。     

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.     

Hadoop平台的性能优化研究

随着基于MapReduce模型的应用程序越来越多,Hadoop性能取决于应用程序。针对上述特性,从应用着手剖析Hadoop存在的局限和不足,提出解决方案,利用作业和任务的多重并发平衡磁盘和网络带宽,减小瓶颈出现的可能性,提高系统性能。     

CloudFlow: A data-aware programming model for cloud workflow applications on modern HPC systems

Traditional High-Performance Computing (HPC) based big-data applications are usually constrained by having to move large amount of data to compute facilities for real-time processing purpose. Modern HPC systems, represented by High-Throughput Computing (HTC) and Many-Task Computing (MTC) platforms, on the other hand, intend to achieve the long-held dream of moving compute to data instead. This kind of data-aware scheduling, typically represented by Hadoop MapReduce, has been successfully implemented in its Map Phase, whereby each Map Task is sent out to the compute node where the corresponding input data chunk is located. However, Hadoop MapReduce limits itself to a one-map-to-one-reduce framework, leading to difficulties for handling complex logics, such as pipelines or workflows. Meanwhile, it lacks built-in support and optimization when the input datasets are shared among multiple applications and/or jobs. The performance can be improved significantly when the knowledge of the shared and frequently accessed data is taken into scheduling decisions.To enhance the capability of managing workflow in modern HPC system, this paper presents CloudFlow, a Hadoop MapReduce based programming model for cloud workflow applications. CloudFlow is built on top of MapReduce, which is proposed not only being data aware, but also shared-data aware. It identifies the most frequently shared data, from both task-level and job-level, replicates them to each compute node for data locality purposes. It also supports user-defined multiple Map- and Reduce functions, allowing users to orchestrate the required data-flow logic. Mathematically, we prove the correctness of the whole scheduling framework by performing theoretical analysis. Further more, experimental evaluation also shows that the execution runtime speedup exceeds 4X compared to traditional MapReduce implementation with a manageable time overhead.     

基于MapReduce的决策树算法并行化

针对传统决策树算法不能解决海量数据挖掘以及ID3算法的多值偏向问题,设计和实现了一种基于MapReduce架构的并行决策树分类算法。该算法采用属性相似度作为测试属性的选择标准来避免ID3算法的多值偏向问题,采用MapReduce模型来解决海量数据挖掘问题。在用普通PC搭建的Hadoop集群的实验结果表明:基于MapReduce的决策树算法可以处理大规模数据的分类问题,具有较好的可扩展性,在保证分类正确率的情况下能获得接近线性的加速比。     

基于决策树挖掘算法的气象大数据云平台设计

大数据、云计算技术的迅猛发展为挖掘气象数据丰富的科研和经济价值提供了技术支撑,促进了Hadoop及其包含的文件存储系统(HDFS,Hadoop Distributed File System)和分布式计算模型在气象数据处理领域广泛应用。由于气象数据具有大数据的4V特征,还需要引入新的数据处理算法来提高气象数据处理效率。通过对决策树算法原理的研究,基于Hadoop云平台,创建随机森林模型,为数据挖掘算法在云平台上的应用提供一种新的可能性。基于决策树(CART,Classification And Regression Trees)挖掘算法的气象大数据云平台设计,采用Hadoop系统架构和MapReduce工作流程,对气象大数据云平台采用集群部署。平台总体架构分为基础设施层、数据管理与处理层、应用层,减少了决策树建立的时间,实现了气象数据高效加工和挖掘分析等平台功能。     

A hadoop based platform for natural language processing of web pages and documents

The rapid and extensive pervasion of information through the web has enhanced the diffusion of a huge amount of unstructured natural language textual resources. A great interest has arisen in the last decade for discovering, accessing and sharing such a vast source of knowledge. For this reason, processing very large data volumes in a reasonable time frame is becoming a major challenge and a crucial requirement for many commercial and research fields. Distributed systems, computer clusters and parallel computing paradigms have been increasingly applied in the recent years, since they introduced significant improvements for computing performance in data-intensive contexts, such as Big Data mining and analysis. Natural Language Processing, and particularly the tasks of text annotation and key feature extraction, is an application area with high computational requirements; therefore, these tasks can significantly benefit of parallel architectures. This paper presents a distributed framework for crawling web documents and running Natural Language Processing tasks in a parallel fashion. The system is based on the Apache Hadoop ecosystem and its parallel programming paradigm, called MapReduce. In the specific, we implemented a MapReduce adaptation of a GATE application and framework (a widely used open source tool for text engineering and NLP). A validation is also offered in using the solution for extracting keywords and keyphrase from web documents in a multi-node Hadoop cluster. Evaluation of performance scalability has been conducted against a real corpus of web pages and documents.     

MapReduce Workload Modeling with Statistical Approach

Large-scale data-intensive cloud computing with the MapReduce framework is becoming pervasive for the core business of many academic, government, and industrial organizations. Hadoop, a state-of-the-art open source project, is by far the most successful realization of MapReduce framework. While MapReduce is easy- to-use, efficient and reliable for data-intensive computations, the excessive configuration parameters in Hadoop impose unexpected challenges on running various workloads with a Hadoop cluster effectively. Consequently, developers who have less experience with the Hadoop configuration system may devote a significant effort to write an application with poor performance, either because they have no idea how these configurations would influence the performance, or because they are not even aware that these configurations exist. There is a pressing need for comprehensive analysis and performance modeling to ease MapReduce application development and guide performance optimization under different Hadoop configurations. In this paper, we propose a statistical analysis approach to identify the relationships among workload characteristics, Hadoop configurations and workload performance. We apply principal component analysis and cluster analysis to 45 different metrics, which derive relationships between workload characteristics and corresponding performance under different Hadoop configurations. Regression models are also constructed that attempt to predict the performance of various workloads under different Hadoop configurations. Several non-intuitive relationships between workload characteristics and performance are revealed through our analysis and the experimental results demonstrate that our regression models accurately predict the performance of MapReduce workloads under different Hadoop configurations.     

Toward objective software process information: experiences from a case study

A critical problem in software development is the monitoring, control and improvement in the processes of software developers. Software processes are often not explicitly modeled, and manuals to support the development work contain abstract guidelines and procedures. Consequently, there are huge differences between ‘actual’ and ‘official’ processes: “the actual process is what you do, with all its omissions, mistakes, and oversights. The official process is what the book, i.e., a quality manual, says you are supposed to do” (Humphrey in A discipline for software engineering. Addison-Wesley, New York, 1995). Software developers lack support to identify, analyze and better understand their processes. Consequently, process improvements are often not based on an in-depth understanding of the ‘actual’ processes, but on organization-wide improvement programs or ad hoc initiatives of individual developers. In this paper, we show that, based on particular data from software development projects, the underlying software development processes can be extracted and that automatically more realistic process models can be constructed. This is called software process mining (Rubin et al. in Process mining framework for software processes. Software process dynamics and agility. Springer Berlin, Heidelberg, 2007). The goal of process mining is to better understand the development processes, to compare constructed process models with the ‘official’ guidelines and procedures in quality manuals and, subsequently, to improve development processes. This paper reports on process mining case studies in a large industrial company in The Netherlands. The subject of the process mining is a particular process: the change control board (CCB) process. The results of process mining are fed back to practice in order to subsequently improve the CCB process.     

MapReduce并行计算技术发展综述

经过几年的发展,并行编程模型MapReduce产生了若干个改进框架,它们都是针对传统MapReduce的不足进行的修正或重写. 本文阐述和分析了这些研究成果,包括： 以HaLoop为代表的迭代计算框架、以Twitter Storm为代表的实时计算框架、以Apache Hama为代表的图计算框架以及以Apache YARN为代表的框架管理平台. 这些专用系统在大数据领域发挥着越来越重要的作用.