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版本中能够准确估算MapReduce作业负载执行时间的性能模型.该模型包括一个优先级树模型与一个排队网络模型,分别用于展示一个MapReduce作业中不同任务之间的依赖关系及MapReduce作业内的同步约束.最后,实验证明了该模型的可用性.     

OPTIMA: On-Line Partitioning Skew Mitigation for MapReduce with Resource Adjustment

Partitioning skew has been shown to be a major issue that can significantly prolong the execution time of MapReduce jobs. Most of the existing off-line heuristics for partitioning skew mitigation are inefficient; they have to wait for the completion of all the map tasks. Some solutions can tackle this problem on-line, but will impose an additional overhead by repartitioning the workload of overloaded tasks. In this paper, we present OPTIMA, an on-line partitioning skew mitigation technique for MapReduce. OPTIMA predicts the workload distribution of reduce tasks at run-time, leverages the deviation detection technique to identify the overloaded tasks and pro-actively adjusts resource allocation for these tasks to reduce their execution time. We provide the upper bound of OPTIMA in time complexity, while allowing OPTIMA to perform totally on-line. Through experiments using both real and synthetic workloads running on an 11-node Hadoop cluster, we have observed OPTIMA can effectively mitigate the partitioning skew and improved the job completion time by up to 36.73 % in our experiments.     

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.     

Hadoop迭代优化技术的研究

Hadoop是处理海量数据的分布式计算框架,已经得到了广泛的应用。但是Hadoop处理图结构数据存在一些不足。图结构数据的强耦合特性,无法通过一次MapReduce计算得出结果,而是需要迭代计算,甚至一次迭代需要多次Ma-pReduce完成。而重新启动MapReduce作业,开销较大,以及迭代过程中可能存在静态数据的不必要传输。文中在Hadoop的基础之上,提出map端存储的策略,即将静态数据存储在map端,在map端完成静态与动态数据相关的计算,减少了整个迭代计算的总运行时间。通过搭建修改过的Hadoop平台,与改进前迭代方案进行比较,实验结果表明map端存储策略运行时间得到了一定程度的减少。     

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.     

HAT: history-based auto-tuning MapReduce in heterogeneous environments

In MapReduce model, a job is divided into a series of map tasks and reduce tasks. The execution time of the job is prolonged by some slow tasks seriously, especially in heterogeneous environments. To finish the slow tasks as soon as possible, current MapReduce schedulers launch a backup task on other nodes for each of the slow tasks. However, traditional MapReduce schedulers cannot detect slow tasks correctly since they cannot estimate the progress of tasks accurately (Hadoop home page http://hadoop.apache.org/, 2011; Zaharia et al. in 8th USENIX symposium on operating systems design and implementation, ACM, New York, pp. 29–42, 2008). To solve this problem, this paper proposes a History-based Auto-Tuning (HAT) MapReduce scheduler, which calculates the progress of tasks accurately and adapts to the continuously varying environment automatically. HAT tunes the weight of each phase of a map task and a reduce task according to the value of them in history tasks and uses the accurate weights of the phases to calculate the progress of current tasks. Based on the accurate-calculated progress of tasks, HAT estimates the remaining time of tasks accurately and further launches backup tasks for the tasks that have the longest remaining time. Experimental results show that HAT can significantly improve the performance of MapReduce applications up to 37% compared with Hadoop and up to 16% compared with LATE scheduler.     

MapReduce:新型的分布式并行计算编程模型

MapReduce是Google提出的分布式并行计算编程模型,用于大规模数据的并行处理。Ma-pReduce模型受函数式编程语言的启发,将大规模数据处理作业拆分成若干个可独立运行的Map任务,分配到不同的机器上去执行,生成某种格式的中间文件,再由若干个Reduce任务合并这些中间文件获得最后的输出文件。用户在使用MapReduce模型进行大规模数据处理时,可以将主要精力放在如何编写Map和Reduce函数上,其它并行计算中的复杂问题诸如分布式文件系统、工作调度、容错、机器间通信等都交给MapReduce系统处理,在很大程度上降低了整个编程难度。MapReduce日益成为云计算平台的主流编程模型。Apache Hadoop项目提供开源的MapReduce系统还有待进一步完善。     

A security framework in G-Hadoop for big data computing across distributed Cloud data centres

MapReduce is regarded as an adequate programming model for large-scale data-intensive applications. The Hadoop framework is a well-known MapReduce implementation that runs the MapReduce tasks on a cluster system. G-Hadoop is an extension of the Hadoop MapReduce framework with the functionality of allowing the MapReduce tasks to run on multiple clusters. However, G-Hadoop simply reuses the user authentication and job submission mechanism of Hadoop, which is designed for a single cluster. This work proposes a new security model for G-Hadoop. The security model is based on several security solutions such as public key cryptography and the SSL protocol, and is dedicatedly designed for distributed environments. This security framework simplifies the users authentication and job submission process of the current G-Hadoop implementation with a single-sign-on approach. In addition, the designed security framework provides a number of different security mechanisms to protect the G-Hadoop system from traditional attacks.     

FF现场总线系统中实时任务的建模与调度

本文研究了单网段 FF现场总线系统中具有时间约束和次序约束的实时任务，即 功能块任务和通信任务的建模与调度．首先，将功能块任务和通信任务等视为相同的任务， 在只考虑任务间次序约束的情况下，提出了基于紧凑模式的任务模型，以保证每个作业被尽 可能早地完成．其次，考虑单网段通信任务共享一个传输介质而引起的通信超时，提出了基 于作业速率单调优先级算法的扩展紧凑模式的任务调度算法，以满足实时任务的时间约束 和次序约束．最后，通过一个应用实例来描述实时任务的调度过程．     

A Task-Based Greedy Scheduling Algorithm for Minimizing Energy of MapReduce Jobs

MapReduce and its open source implementation, Hadoop, have gained widespread adoption for parallel processing of big data jobs. Since the number of such big data jobs is also rapidly rising, reducing their energy consumption is increasingly more important to reduce environmental impact as well as operational costs. Prior work by Mashayekhy et al. (IEEE Trans. Parallel Distributed Syst. 26, 2720–2733, 2016), has tackled the problem of energy-aware scheduling of a single MapReduce job but we provide a far more efficient heuristic in this paper. We first model the problem as an Integer Linear Program to find the optimal solution using ILP solvers. Then we present a task-based greedy scheduling algorithm, TGSAVE, to select a slot for each task to minimize the total energy consumption of the MapReduce job for big data applications in heterogeneous environments without significant performance loss while satisfying the service level agreement (SLA). We perform several experiments on a Hadoop cluster to measure characteristics of tasks for nine different applications to evaluate our proposed algorithm. The results show that the total energy consumption of MapReduce jobs obtained by TGSAVE is up to 35% less than that achieved by EMRSA proposed in Mashayekhy et al. (IEEE Trans. Parallel Distributed Syst. 26, 2720–2733, 2016), its closest rival, for same workloads. Besides, TGSAVE is capable of finding a solution in same order of time for up to 74% tighter deadlines than the tightest deadline that EMRSA can find a feasible one. On average, TGSAVE solution is approximately 1.4% far from the optimal solution, and it can meet deadlines as tight as 12%, on average, above the energy-oblivious minimum makespan in the benchmarks we examined.     

Handling partitioning skew in MapReduce using LEEN

MapReduce is emerging as a prominent tool for big data processing. Data locality is a key feature in MapReduce that is extensively leveraged in data-intensive cloud systems: it avoids network saturation when processing large amounts of data by co-allocating computation and data storage, particularly for the map phase. However, our studies with Hadoop, a widely used MapReduce implementation, demonstrate that the presence of partitioning skew (Partitioning skew refers to the case when a variation in either the intermediate keys’ frequencies or their distributions or both among different data nodes) causes a huge amount of data transfer during the shuffle phase and leads to significant unfairness on the reduce input among different data nodes. As a result, the applications severe performance degradation due to the long data transfer during the shuffle phase along with the computation skew, particularly in reduce phase. In this paper, we develop a novel algorithm named LEEN for locality-aware and fairness-aware key partitioning in MapReduce. LEEN embraces an asynchronous map and reduce scheme. All buffered intermediate keys are partitioned according to their frequencies and the fairness of the expected data distribution after the shuffle phase. We have integrated LEEN into Hadoop. Our experiments demonstrate that LEEN can efficiently achieve higher locality and reduce the amount of shuffled data. More importantly, LEEN guarantees fair distribution of the reduce inputs. As a result, LEEN achieves a performance improvement of up to 45 % on different workloads.     

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.     

异构大数据编程环境 Hadoop+

互联网和物联网技术的飞速发展开启了“大数据”时代。目前,硬件的高速发展催生了许多异构芯片,它们越来越多地出现在大规模数据中心里,支持不同的应用程序,在提升性能的同时降低整体功耗。文章重点介绍了基于 MapReduce编程模型的 Hadoop+框架的设计与实现,它允许用户在单个任务中调用 CUDA/OpenCL的并行实现,并能通过异构任务模型帮助用户。在我们的实验平台上,五种常见机器学习算法使用 Hadoop+框架相对于 Hadoop能达到1.4×～16.1×的加速比,在 Hadoop+框架中使用异构任务模型指导其资源分配策略,对单个应用负载上最高达到36.0%的性能提升;对多应用的混合负载,最多能减少36.9%,平均17.6%的应用执行时间。     

基于输入分片扰乱的BP神经网络MapReduce训练方法

BP神经网络的MapReduce训练中,每个map训练任务产生的中间权阵只对该训练节点上的输入分片收敛,为提高BP神经网络的训练效率,保证MapReduce训练的全局收敛性,提出一种基于输入分片扰乱的MapReduce训练方法。通过对训练样本集进行系统抽样来扰乱输入分片,并产生新的输入分片,依靠新的输入分片以map任务的原权阵为基础进行迭代训练,可加速MapReduce训练达到收敛的进程;为提高map训练任务的局部收敛速度,在每轮次的训练完成后,选取map任务产生的权阵中全局误差最小者,作为下轮次各map训练任务的初始权阵。在Hadoop集群上的实验表明,该方法可使MapReduce训练BP神经网络的效率得到很大提升。     

Improving MapReduce Performance with Partial Speculative Execution

The MapReduce framework has become the de facto standard for big data processing due to its attractive features and abilities. One is that it automatically parallelizes a job into multiple tasks and transparently handles task execution on a large cluster of commodity machines. The increasing heterogeneity of distributed environments may result in a few straggling tasks, which prolong job completion. Speculative execution is proposed to mitigate stragglers. However, the existing speculative execution mechanism could not work efficiently as many speculative tasks are still slower than their original tasks. In this paper, we explore an approach to increase the efficiency of speculative execution, and further improve MapReduce performance. We propose the Partial Speculative Execution (PSE) strategy to make speculative tasks start from the checkpoint. By leveraging the checkpoint of original tasks, PSE can eliminate the costs of re-reading, re-copying, and re-computing the processed data. We implement PSE in Hadoop, and evaluate its performance in terms of job completion time and the efficiency of speculative execution under several kinds of classical workloads. Experimental results show that, in heterogeneous environments with stragglers, PSE completes jobs 56 % faster than that with no speculation and 12 % faster than that with LATE, an improved speculative execution algorithm. In addition, on average PSE can improve the efficiency of speculative execution by 24 % compared to LATE.     

基于排队网络的多优先级MapReduce作业调度算法

MapReduce是一个能够对大规模数据进行分布式处理的框架,目前被各个领域广泛应用。在提供MapReduce服务的集群中,如何保证不同优先级用户的截止时间限定是MapReduce作业调度问题的一个挑战。针对这一问题,提出了一个基于排队网络的多优先级作业调度算法（MPSA）。首先分析和归纳了基于MapReduce模型的算法,提出了三种常见模式,采用Jackson排队网络对基于MapReduce模型的算法建立了数学模型,应用该网络模型可以求出不同优先级队列对资源的需求;随后使用AR(1)模型进行预测,使算法可以动态地适应不同的用户访问量;利用二分查找算法,分步计算出不同优先级在map阶段和reduce阶段分配的槽位数;最后实现了在MapReduce模型中应用的实时调度算法。实验结果表明,与传统的FIFO和公平调度算法相比,本文提出的算法在用户到达率和任务规模变化的情况下,可以更加有效地满足不同优先级用户的截止时间限定。     

基于MapReduce的分布式AP聚类算法

随着网络的普遍应用,网络中产生的数据急剧增长,大规模数据处理面临严峻挑战。本文在对AP聚类算法进行研究的基础上,利用MapReduce编程模型思想对AP聚类算法进行改进,设计在云平台Hadoop环境下运行的基于MapReduce的分布式AP聚类算法,并在实验中对不同规模的图数据进行聚类测试,实验结果表明分布式的AP聚类算法具有很好的时间效率和加速比。      

混合存储模式下MapReduce作业调度

在异构Hadoop集群场景中, 为了缓和由于纠删码和副本存储模式混合使用, 以及服务器节点本身实时算力差异造成的MapReduce作业处理效率低下的问题, 本文实现了一种根据数据存储情况和节点实时负载来在多并发场景下动态调节MapReduce作业任务分配情况的调度策略. 该策略通过修改当前Hadoop框架中的数据存储选址策略并对节点任务并发量进行动态控制, 在多作业并发时实现更加均衡的作业间资源分配. 实验结果表明, 相较于Hadoop默认的两种作业调度策略, 本文提出的调度模式能够将作业完成时间缩短约17%, 并有效避免部分作业面临的饥饿现象.     

The HaLoop approach to large-scale iterative data analysis

The growing demand for large-scale data mining and data analysis applications has led both industry and academia to design new types of highly scalable data-intensive computing platforms. MapReduce has enjoyed particular success. However, MapReduce lacks built-in support for iterative programs, which arise naturally in many applications including data mining, web ranking, graph analysis, and model fitting. This paper (This is an extended version of the VLDB 2010 paper “HaLoop: Efficient Iterative Data Processing on Large Clusters” PVLDB 3(1):285–296, 2010.) presents HaLoop, a modified version of the Hadoop MapReduce framework, that is designed to serve these applications. HaLoop allows iterative applications to be assembled from existing Hadoop programs without modification, and significantly improves their efficiency by providing inter-iteration caching mechanisms and a loop-aware scheduler to exploit these caches. HaLoop retains the fault-tolerance properties of MapReduce through automatic cache recovery and task re-execution. We evaluated HaLoop on a variety of real applications and real datasets. Compared with Hadoop, on average, HaLoop improved runtimes by a factor of 1.85 and shuffled only 4 % as much data between mappers and reducers in the applications that we tested.