面向并行负载平衡的数据剖分技术*

对传统的数据剖分技术和负载平衡对大规模并行计算性能的影响进行了综述,介绍了目前典型的几何剖分方法和图剖分方法的特点,并分析比较各种剖分算法及常用剖分软件包(ParMETIS、Zoltan、JOSTLE等)在实际应用中的优缺点,深入探讨了数据剖分技术是如何对超大规模数值模拟计算任务进行高效划分以解决负载平衡问题的,以期为开展并行计算研究和并行性能优化的研究人员提供参考。     

On-line load balancing of temporary tasks revisited

We study load balancing problems of temporary jobs (i.e., jobs that arrive and depart at unpredictable time) in two different contexts, namely, machines and network paths. Such problems are known as machine load balancing and virtual circuit routing in the literature. We present new on-line algorithms and improved lower bounds.     

On runtime parallel scheduling for processor load balancing

Parallel scheduling is a new approach for load balancing. In parallel scheduling, all processors cooperate to schedule work. Parallel scheduling is able to accurately balance the load by using global load information at compile-time or runtime. It provides high-quality load balancing. This paper presents an overview of the parallel scheduling technique. Scheduling algorithms for tree, hypercube, and mesh networks are presented. These algorithms can fully balance the load and maximize locality at runtime. Communication costs are significantly reduced compared to other existing algorithms     

Online vector scheduling and generalized load balancing

We give a polynomial time reduction from the vector scheduling problem (VS) to the generalized load balancing problem (GLB). This reduction gives the first non-trivial online algorithm for VS where vectors come in an online fashion. The online algorithm is very simple in that each vector only needs to minimize the _Lln(md)$L_{ln\left(md\right)}$ norm of the resulting load when it comes, where m$m$ is the number of partitions and d$d$ is the dimension of vectors. It has an approximation bound of elog(md)$elog\left(md\right)$, which is in O(ln(md))$O(ln\left(md\right))$, so it also improves the O(ln²d)$O\left({ln}^{2}d\right)$ bound of the existing polynomial time algorithm for VS. Additionally, the reduction shows that GLB does not have constant approximation algorithms that run in polynomial time unless P=NP$P=NP$.     

P2P分域网络动态负载均衡调度的抗脆弱性策略

P2P网络中集散节点的存在加重了网络的脆弱性。提出一种动态负载均衡调度的抗脆弱性策略,该策略依据节点物理位置的邻近性对网络进行了分域,并设计出更为公平的函数对节点的负载进行动态的量化,当消费节点发出下载请求后,选择负载动态变化因子最小的节点作为服务节点,避免了单个节点负载猛增的情况,均衡了节点的负载。仿真证明该策略能有效抑制网络中集散节点的形成,增强网络的健壮性,从而达到保障P2P网络可持续健康发展的目的。     

基于Q值法的负载均衡调度算法研究

随着网络的普及,大型网络游戏的开发,网络超载与超负荷时常发生,如何在调度算法层面上实质性的改变这一状况成为当务之急。本文将具体介绍一种基于Q值法的负载均衡调度算法改变这一状况,该算法在国内属首创。     

一种异构云环境下的负载均衡算法

针对动态负载均衡算法在异构云环境中的任务迁移次数过多的问题,提出了一种最小化任务迁移次数的动态负载均衡(MMLB)算法。MMLB算法通过自适应阈值对虚拟机进行分组、任务选择算法最小化任务迁移的次数、任务调度算法优化任务分配实现了任务的再分配。将MMLB与WRR、HBBLB、LBF算法进行实验对比分析,MMLB算法在Makespan、平均任务响应时间、负载不均衡度等评价指标上表现更优,并且有效降低了任务迁移的次数。实验结果验证了MMLB算法的可行性和有效性。     

Real-time load balancing scheduling algorithm for periodic simulation models

A scheduling algorithm is crucial for real-time simulations because it guarantees that each model meets its deadline. Traditional online real-time scheduling algorithms such as Earliest Deadline First (EDF) introduce a high overhead when scheduling a large number of models. In this paper, a new algorithm called time-stepped load balancing (TLS) is proposed to address the real-time execution of a model set in a time-stepped simulation. A load balancing schedule table is generated before a simulation and rebalanced at runtime to dynamically schedule the changed model set. This table is organized by the execution periods of the models and balanced according to the load of each time step. Moreover, the slack time is distributed evenly among the steps to improve the real-time reliability. An extension to the algorithm for a multi-core environment is further studied to address those models with long execution times. Experimental results show that our scheduling algorithm outperforms the classical EDF approach. The highest performance improvement of TLS over EDF reaches 3–4% in terms of saving processor resources, and the jitter is about 4 times less when 90 entities are employed in a typical tank combat simulation scenario.     

异构云中面向集群负载均衡的任务调度策略

负载均衡是提高资源利用率和系统稳定性的重要手段。基于改进的自适应变异粒子群算法,提出了一种异构环境下面向集群负载均衡的任务调度策略。在调度策略的设计中,融入了经济学“二八”定律,通过把握用户对集群节点安全性和可靠性的偏好程度并预估任务的负载信息,在保证系统负载尽量均衡的前提下,最小化任务执行时间的同时提高大客户满意度。仿真实验显示,改进的自适应变异粒子群算法比未改进的自适应变异粒子群算法和基本粒子群算法在收敛速度和跳出局部最优两个方面都有更好的表现。结果表明,改进的自适应变异粒子群算法在保证集群负载均衡的同时可以更好地提高云服务提供商的利润空间。     

负载平衡调度问题的一般模型研究

分布式系统提供了巨大的处理能力,为了实现和充分利用这种能力,需要优良的负载平衡调度技术。因此,负载平衡问题是影响分布式系统性能的重要因素。在深入研究分布式系统中负载平衡调度问题的基础上,归纳总结了负载平衡调度的一般模型,对影响负载平衡的各个因素进行了详细的分析。此模型已在一个实际模型中得到了有效地验证。     

Task scheduling,resource provisioning,and load balancing on scientific workflows using parallel SARSA reinforcement learning agents and genetic algorithm

The Journal of Supercomputing - Cloud computing is one of the most popular distributed environments, in which, multiple powerful and heterogeneous resources are used by different user applications....     

Data partitioning and load balancing in parallel disk systems

Parallel disk systems provide opportunities for exploiting I/O parallelism in two possible ways, namely via inter-request and intra-request parallelism. In this paper, we discuss the main issues in performance tuning of such systems, namely striping and load balancing, and show their relationship to response time and throughput. We outline the main components of an intelligent, self-reliant file system that aims to optimize striping by taking into account the requirements of the applications, and performs load balancing by judicious file allocation and dynamic redistributions of the data when access patterns change. Our system uses simple but effective heuristics that incur only little overhead. We present performance experiments based on synthetic workloads and real-life traces. Received May 17, 1994 / Accepted June 9, 1997     

基于动态负载均衡策略的网格任务调度优化模型和算法

任务调度是一个NP-hard问题,而且是并行与分布式计算中一个必不可少的组成部分,特别是在网格计算环境中任务调度更加复杂。结合免疫克隆算法和模拟退火算法的优点,提出了一种网格任务调度优化模型和算法。仿真实验结果表明,这种调度算法有效地实现了资源的负载均衡,克服了遗传算法容易陷入局部最优的缺点,可以成功地应用于网格任务调度中。     

面向云环境的Flink负载均衡策略

作为新一代的大数据计算引擎,Flink得到了广泛应用。Flink在云环境下进行容器化部署时,其默认任务调度算法不能感知节点的资源信息,导致即时调整负载和自主均衡能力较差,而主流的容器编排工具虽然提供了管理容器的可能性,却也未能结合Flink特点解决平衡资源利用的同时降低容器组内的通信开销问题。针对以上问题开展研究,提出了一种面向云环境的Flink负载均衡策略FLBS,综合考虑了Flink集群中算子的分布特点和容器间通信机制,以节点间通信开销和均衡负载作为评估标准。实验结果表明,与Flink默认调度策略相比,FLBS能够有效提高计算效率,提升系统性能。     

Resource scheduling algorithm with load balancing for cloud service provisioning

Cloud computing uses scheduling and load balancing for virtualized file sharing in cloud infrastructure. These two have to be performed in an optimized manner in cloud computing environment to achieve optimal file sharing. Recently, Scalable traffic management has been developed in cloud data centers for traffic load balancing and quality of service provisioning. However, latency reducing during multidimensional resource allocation still remains a challenge. Hence, there necessitates efficient resource scheduling for ensuring load optimization in cloud. The objective of this work is to introduce an integrated resource scheduling and load balancing algorithm for efficient cloud service provisioning. The method constructs a Fuzzy-based Multidimensional Resource Scheduling model to obtain resource scheduling efficiency in cloud infrastructure. Increasing utilization of Virtual Machines through effective and fair load balancing is then achieved by dynamically selecting a request from a class using Multidimensional Queuing Load Optimization algorithm. A load balancing algorithm is then implemented to avoid underutilization and overutilization of resources, improving latency time for each class of request. Simulations were conducted to evaluate the effectiveness using Cloudsim simulator in cloud data centers and results shows that the proposed method achieves better performance in terms of average success rate, resource scheduling efficiency and response time. Simulation analysis shows that the method improves the resource scheduling efficiency by 7% and also reduces the response time by 35.5 % when compared to the state-of-the-art works.     

Distributed scheduling strategy for divisible loads on arbitrarily configured distributed networks using load balancing via virtual routing

In this paper, we consider a scheduling problem for divisible loads originating from single or multiple sites on arbitrary networks. We first propose a generalized mathematical model and formulate the scheduling problem as an optimization problem with an objective to minimize the processing time of the loads. We derive a number of theoretical results on the solution of the optimization problem. On the basis of these first set of results, we propose an efficient algorithm for scheduling divisible loads using the concept of load balancing via virtual routing for an arbitrary network configuration. The proposed algorithm has three major attractive features. Firstly, the algorithm is simple to realize and can be implemented in a distributed fashion. The second one is in its style of working by avoiding the need for generating a timing diagram explicitly for any complex networks having an arbitrary network topology. The last one is its capability of handling divisible loads originating from both single and multiple sites. When divisible loads originate from a single node, we compare the proposed algorithm with a recently proposed RAOLD algorithm which is based on minimum cost spanning tree [J. Yao, V. Bharadwaj, Design and performance analysis of divisible load scheduling strategies on arbitrary graphs, Cluster Computing 7(2) (2004) 191–207]. When divisible loads originate from multiple sites, we test the performance on sparse, medium and densely connected networks. This is the first time in the divisible load theory (DLT) literature that such a generic approach for handling divisible loads originating from multiple sites on arbitrary networks employing load balancing via virtual routing is attempted.     

面向软件即服务的负载均衡策略建模与分析

为提高软件即服务（SaaS）应用中资源的访问效率,提出支持SaaS服务重要特征的负载均衡策略。首先,结合SaaS服务的多租户和高度可伸缩两大特性,提出一种基于租户请求分流、在局部和全局两个层次伸缩的负载均衡策略;其次,对所提出负载均衡策略用Petri网进行建模并仿真;最后,将提出的负载均衡策略与轮询（RR）、随机和改进的最小连接（ILCS）负载均衡算法在总体响应时间和总吞吐量两方面进行比较。实验结果表明：在请求速率达到500请求/秒后,所提策略的总体响应时间和总吞吐量趋于稳定并优于另外三种算法。     

基于通信负载均衡的社交网络图分割方法

海量社交网络数据中蕴含着丰富的信息,图论是挖掘这些信息的重要方法之一。面对日益增多的图数据,分布式计算成为处理大规模图数据的有效手段。在分布式图计算中,通信所消耗的时间占有很大的比例,通过图分割算法的设计可以有效地降低通信量并实现负载均衡,从而提高分布式图计算的效率,典型的例子包括Metis图分割算法。但是,用现有的图分割算法处理非均衡图数据会造成各个子图之间通信量不均衡,从而影响了计算效率。为了解决这一问题,提出一种新的图分割方法：通信均衡标签交换方法。该方法在保持子图规模一致的基础上,既降低了全图计算所需的通信量,又使各个子图之间的通信量达到均衡。实验结果表明,与Metis等典型的图分割算法相比,提出的图分割方法在各种数据集和集群配置情况下,能降低6%~30%的图计算时间,充分显示了该方法的有效性。     

Energy-aware mixed partitioning scheduling in standby-sparing systems

Previous standby-sparing techniques assume that all tasks don't access to shared resources. In addition, primary tasks and backup tasks are allocated to the primary processor and spare processor respectively. Spare processor schedules tasks with maximum processor speed. Unlike previous techniques, we have studied the problem of minimizing energy consumption and preserving the original reliability for dynamic-priority real-time task set with shared resources in a standby-sparing system. We propose a novel energy-aware mixed partitioning scheduling algorithm (EAMPSA). Earliest deadline first/dynamic deadline modification (EDF/DDM) scheduling scheme is used to ensure that the shared resources can be accessed in a mutual exclusive manner. Uniformly speed is used to the primary processor and the spare processor. In addition, we use the mixed mapping partitioning of primary and backup tasks method to map tasks. A novel method of mapping task is proposed i.e. the tasks which need to access to shared resources are mapped into the primary processor and the tasks which have no resource requirements are mapped into the spare processor. Furthermore, DVS and DPM techniques are used for both primary and backup tasks to save energy. The experimental results show that the EAMPSA algorithm consumes average 55.43% less energy than that of the SSPT algorithm.