Performance-based dynamic loop scheduling in heterogeneous computing environments

With the rapid advance of computing technologies, it becomes more and more common to construct high-performance computing environments with heterogeneous commodity computers. Previous loop scheduling schemes were not designed for this kind of environments. Therefore, better loop scheduling schemes are needed to further increase the performance of the emerging heterogeneous PC cluster environments. In this paper, we propose a new heuristic for the performance-based approach to partition loop iterations according to the performance weighting of cluster/grid nodes. In particular, a new parameter is proposed to consider HPCC benchmark results as part of performance estimation. A heterogeneous cluster and grid were built to verify the proposed approach, and three kinds of application program were implemented for execution on cluster testbed. Experimental results show that the proposed approach performs better than the previous schemes on heterogeneous computing environments.     

An Enhanced Parallel Loop Self-Scheduling Scheme for Cluster Environments

Approaches for dealing with scheduling and load-balancing in PC-based cluster systems are famous and well known. In such environments, Self-Scheduling Schemes are suitable for parallel loops with independent iterations. However, while schemes such as FSS, GSS, and TSS fit most computer systems, they cannot provide good load-balancing. Chao-Tung Yang and Shun-Chi Chang proposed a parallel loop scheduling scheme for heterogeneous PC cluster systems in Yang and Chang [13]. Though the proposed scheme allows users to choose parameters before execution initialization, weaknesses in it motivated us to develop further improvements. For instance, using fixed and monotonous parameters can easily lead to invalid scheduling due to use of previously input information. Thus, in this paper we propose a new scheme that fits most widely available computer systems and allows the scheduling parameter to be adjusted dynamically in order to provide higher overall performance.     

Performance‐based parallel loop self‐scheduling using hybrid OpenMP and MPI programming on multicore SMP clusters

Parallel loop self‐scheduling on parallel and distributed systems has been a critical problem and it is becoming more difficult to deal with in the emerging heterogeneous cluster computing environments. In the past, some self‐scheduling schemes have been proposed as applicable to heterogeneous cluster computing environments. In recent years, multicore computers have been widely included in cluster systems. However, previous researches into parallel loop self‐scheduling did not consider certain aspects of multicore computers; for example, it is more appropriate for shared‐memory multiprocessors to adopt Open Multi‐Processing (OpenMP) for parallel programming. In this paper, we propose a performance‐based approach using hybrid OpenMP and MPI parallel programming, which partition loop iterations according to the performance weighting of multicore nodes in a cluster. Because iterations assigned to one MPI process are processed in parallel by OpenMP threads run by the processor cores in the same computational node, the number of loop iterations allocated to one computational node at each scheduling step depends on the number of processor cores in that node. Experimental results show that the proposed approach performs better than previous schemes. Copyright © 2010 John Wiley & Sons, Ltd.     

一种基于内存服务的内存共享网格系统

内存密集型应用对运行环境的物理内存要求严格,在物理内存不足时将会引发大量磁盘IO,降低系统性能．传统的网络内存致力于在集群内部通过共享空闲节点的物理内存解决该问题,但受集群负载和内部网络影响较大．通过结合网络内存和服务计算、网格计算等技术,提出一种基于内存服务的内存共享网格系统——内存网格,并分析和讨论了实现内存服务的关键技术和算法．内存网格弥补了网络内存的不足,扩展了网格计算的应用范围．通过基于真实应用运行状态的模拟,证明了内存网格与网络内存相比具有性能的提高．     

Improving job scheduling performance with parallel access to replicas in Data Grid environment

Data Grid has evolved to be the solution for data-intensive applications, such as High Energy Physics (HEP), astrophysics, and computational genomics. These applications usually have large input of data to be analyzed and these input data are widely replicated across Data Grid to improve the performance. The job scheduling performance on traditional computing jobs can be studied using queuing theory. However, with the addition of data transfer, the job scheduling performance is too complex to be modeled. In this research, we study the impact of data transfer on the performance of job scheduling in the Data Grid environment. We have proposed a parallel downloading system that supports replicating data fragments and parallel downloading of replicated data fragments, to improve the job scheduling performance. The performance of the parallel downloading system is compared with non-parallel downloading system, using three scheduling heuristics: Shortest Turnaround Time (STT), Least Relative Load (LRL) and Data Present (DP). Our simulation results show that the proposed parallel download approach greatly improves the Data Grid performance for all three scheduling algorithms, in terms of the geometric mean of job turnaround time. The advantage of parallel downloading system is most evident when the Data Grid has relatively low network bandwidth and relatively high computing power.     

Reputation-based dependable scheduling of workflow applications in Peer-to-Peer Grids

Grids facilitate creation of wide-area collaborative environment for sharing computing or storage resources and various applications. Inter-connecting distributed Grid sites through peer-to-peer routing and information dissemination structure (also known as Peer-to-Peer Grids) is essential to avoid the problems of scheduling efficiency bottleneck and single point of failure in the centralized or hierarchical scheduling approaches. On the other hand, uncertainty and unreliability are facts in distributed infrastructures such as Peer-to-Peer Grids, which are triggered by multiple factors including scale, dynamism, failures, and incomplete global knowledge.In this paper, a reputation-based Grid workflow scheduling technique is proposed to counter the effect of inherent unreliability and temporal characteristics of computing resources in large scale, decentralized Peer-to-Peer Grid environments. The proposed approach builds upon structured peer-to-peer indexing and networking techniques to create a scalable wide-area overlay of Grid sites for supporting dependable scheduling of applications. The scheduling algorithm considers reliability of a Grid resource as a statistical property, which is globally computed in the decentralized Grid overlay based on dynamic feedbacks or reputation scores assigned by individual service consumers mediated via Grid resource brokers. The proposed algorithm dynamically adapts to changing resource conditions and offers significant performance gains as compared to traditional approaches in the event of unsuccessful job execution or resource failure. The results evaluated through an extensive trace driven simulation show that our scheduling technique can reduce the makespan up to 50% and successfully isolate the failure-prone resources from the system.     

QoS guided Min-Min heuristic for grid task scheduling

Task scheduling is an integrated component of computing.With the emergence of Grid and ubiquitous computing,new challenges appear in task scheduling based on properties such as security,quality of service,and lack of central control within distributed administrative domains.A Grid task scheduling framework must be able to deal with these issues.One of the goals of Grid task scheduling is to achivev high system throughput while matching applications with the available computing resources.This matching of resources in a non-deterministically shared heterogeneous environment leads to concerns over Quality of Service (QoS).In this paper a novel QoS guided task scheduling algorithm for Grid computing is introduced.The proposed novel algorithm is based on a general adaptive scheduling heuristics that includes QoS guidance.The algorithm is evaluated within a simulated Grid environment.The experimental results show that the nwe QoS guided Min-Min heuristic can lead to significant performance gain for a variety of applications.The approach is compared with others based on the quality of the prediction formulated by inaccurate information.     

一种基于人工免疫的新的网格资源调度算法

由于广域网性能的巨大提高和功能强大且价格低廉的计算机不断增多,网格计算以一种极具有前途和吸引力的新范式出现。网格计算是集成地理位置分布,异构,多领域资源的一种平台,它提供透明、安全、同等、高性能资源共享。要获取计算网格中潜在的能量,设计一种有效和高效的网格资源调度算法很重要。网格独特的特点使得网格环境下的资源调度是相当复杂的。本文将重点设计一种新的基于免疫算法的网格资源调度算法。     

一种冗余调度的可靠网格计算模型

为了解决动态、不稳定的网格环境下的可靠计算问题,提出一种基于冗余调度的可靠网格计算模型.首先给出计算网格系统可靠性的定义,并基于系统可靠性定义给出了冗余调度的可靠网格计算模型,设计了冗余调度算法,模拟实验结果证明了提出的模型可以提高计算网格任务调度的可靠性.为了使提出的模型更好应用于实际网格计算环境,给出基于概率的冗余度优化公式,将该公式引入到冗余调度模型,可以获得优化的调度冗余度,不仅可以提高任务调度系统的可靠性,而且能提高资源的利用率.     

Integration of task abortion and security requirements in GA-based meta-heuristics for independent batch grid scheduling

Scheduling and resource allocation in large scale distributed environments, such as Computational Grids (CGs), arise new requirements and challenges not considered in traditional distributed computing environments. Among these new requirements, task abortion and security become needful criteria for Grid schedulers. The former arises due to the dynamics of the Grid systems, in which resources are expected to enter and leave the system in an unpredictable way. The latter requirement appears crucial in Grid systems mainly due to a multi-domain nature of CGs. The main aim of this paper is to develop a scheduling model that enables the aggregation of task abortion and security requirements as additional, together with makespan and flowtime, scheduling criteria into a cumulative objective function. We demonstrate the high effectiveness of genetic-based schedulers in finding near-optimal solutions for multi-objective scheduling problem, where all criteria (objectives) are simultaneously optimized. The proposed meta-heuristics are experimentally evaluated in static and dynamic Grid scenarios by using a Grid simulator. The obtained results show the fast reduction of the values of basic scheduler performance metrics, especially in the dynamic case, that confirms the usefulness of the proposed approach in real-life scenarios.     

G‐BLAST: a Grid‐based solution for mpiBLAST on computational Grids

Over the past few years, research and development in bioinformatics (e.g. genomic sequence alignment) has grown with each passing day fueling continuing demands for vast computing power to support better performance. This trend usually requires solutions involving parallel computing techniques because cluster computing technology reduces execution times and increases genomic sequence alignment efficiency. One example, mpiBLAST is a parallel version of NCBI BLAST that combines NCBI BLAST with message passing interface (MPI) standards. However, as most laboratories cannot build up powerful cluster computing environments, Grid computing framework concepts have been designed to meet the need. Grid computing environments coordinate the resources of distributed virtual organizations and satisfy the various computational demands of bioinformatics applications. In this paper, we report on designing and implementing a BioGrid framework, called G‐BLAST, that performs genomic sequence alignments using Grid computing environments and accessible mpiBLAST applications. G‐BLAST is also suitable for cluster computing environments with a server node and several client nodes. G‐BLAST is able to select the most appropriate work nodes, dynamically fragment genomic databases, and self‐adjust according to performance data. To enhance G‐BLAST capability and usability, we also employ a WSRF Grid Service Portal and a Grid Service GUI desk application for general users to submit jobs and host administrators to maintain work nodes. Copyright © 2008 John Wiley & Sons, Ltd.     

Online algorithms for advance resource reservations

We consider the problem of providing QoS guarantees to Grid users through advance reservation of resources. Advance reservation mechanisms provide the ability to allocate resources to users based on agreed-upon QoS requirements and increase the predictability of a Grid system, yet incorporating such mechanisms into current Grid environments has proven to be a challenging task due to the resulting resource fragmentation. We use concepts from computational geometry to present a framework for tackling the resource fragmentation, and for formulating a suite of scheduling strategies. We also develop efficient implementations of the scheduling algorithms that scale to large Grids. We conduct a comprehensive performance evaluation study using simulation, and we present numerical results to demonstrate that our strategies perform well across several metrics that reflect both user- and system-specific goals. Our main contribution is a timely, practical, and efficient solution to the problem of scheduling resources in emerging on-demand computing environments.     

EMPEROR: An OGSA Grid Meta-Scheduler Based on Dynamic Resource Predictions

Scheduling constitutes an integral feature of Grid computing infrastructures, being also a key to realizing several of the Grid promises. In particular, scheduling can maximize the resources available to end users, accelerate the execution of jobs, while also supporting scalable and autonomic management of the resources comprising a Grid. Grid scheduling functionality hinges on middleware components called meta-schedulers, which undertake to automatically distribute jobs across the dispersed heterogeneous resources of a Grid. In this paper we present the design and implementation of a Grid meta-scheduler, which we call EMPEROR. EMPEROR provides a framework for implementing scheduling algorithms based on performance criteria. In implementing a particular instantiation of this framework, we have devised models for predicting host load and memory resources, and accordingly for estimating the running time of a task. These models hinge on time series analysis techniques and take into account results of the cluster computing literature. Apart from incorporating these models, EMPEROR provides fully fledged Grid scheduling functionality, which complies with OGSA standards as the later are reflected in the Globus toolkit. Specifically, EMPEROR interfaces to Globus middleware services (i.e., GSI, MDS, GRAM) towards discovering resources, implementing the scheduling algorithm and ultimately submitting jobs to local scheduling systems. By and large, EMPEROR is one of the few standards based meta-schedulers making use of dynamic scheduling information.     

Integrating Computing Resources on Multiple Grid-Enabled Job Scheduling Systems Through a Grid RPC System

We present a framework for a parallel programming model by remote procedure calls, which bridge large-scale computing resource pools managed by multiple Grid-enabled job scheduling systems. With this system, the user can exploit not only remote servers and clusters, but also the computing resources provided by Grid-enabled job scheduling systems located on different sites. This framework requires a Grid remote procedure call (RPC) system to decouple the computation in a remote node from the Grid RPC mechanism and uses document-based communication rather than connection-based communication. We implemented the proposed framework as an extension of the OmniRPC system, which is a Grid RPC system for parallel programming. We designed a general interface to easily adapt the OmniRPC system to various Grid-enabled job scheduling systems, including XtremWeb, CyberGRIP, Condor and Grid Engine. We show the preliminary performance of these implementations using a phylogenetic application. We found that the proposed system can achieve approximately the same performance as OmniRPC and can handle interruptions in worker programs on remote nodes. Yoshihiro Nakajima is a Research Fellow of the Japan Society for the Promotion of Science     

Entropic Grid Scheduling

Computational Grids (CGs) are large scale dynamical networks of geographically distributed peer resource clusters. These clusters are independent but cooperating computing systems bound by a management framework for the provision of computing services, called Grid Services. In its basic form, the Grid scheduling problem consists in finding at least one cluster that has the capacity to handle, within the constraints of a specified quality of service, a user service request submitted to the CG. Since CGs span distinct management domains, the scheduling process has to be decentralized. Furthermore, it has to account for the ubiquitous uncertainty on the state of the CG. In this paper, we propose a scalable distributed Entropy-based scheduling approach that utilizes a Markov chain model to capture the dynamics of the service capacity state. An entropy-based quantification of the uncertainty on the service capacity information is developed and explicitly integrated within the proposed Grid scheduling approach. The performance of the proposed scheduling strategy is validated, through simulation, against a random delegation scheme and a load balancing-based scheduling strategy with respect to throughput, exploitation and convergence speed, respectively.     

The design and implementation of Visuel performance monitoring and analysis toolkit for cluster and grid environments

The computing power provided by high performance and low-cost PC-based clusters with Grid platforms are attractive and they are equal or superior to supercomputers and mainframes. In this paper, we present implementation and design rationale of Visuel toolkit for MPI parallel program performance measurement and analysis in cluster and grid environments. Most of performance visualization tools available today for high-performance platforms show solely system performance data (e.g., CPU load, memory usage, network bandwidth, server average load), and thus, being suitable for computing system activity visualization. The Visuel (Visuel (in French language) = to visualize) toolkit is web-based interface designed to show performance activities of all computing nodes of a distributed environment involved in the execution of MPI parallel program, such as CPU load level and memory usage of each computing node. In addition, this toolkit is able to display comparative performance data charts of MPI parallel applications and multiple executions under investigation. The usage of this toolkit shows that it outperforms in easing the process of investigation of parallel applications.

Scheduling real-time divisible loads with advance reservations

Providing QoS and performance guarantee to arbitrarily divisible loads has become a significant problem for many cluster-based research computing facilities. While progress is being made in scheduling arbitrarily divisible loads, previous approaches have no support for advance reservations. However, with the emergence of Grid applications that require simultaneous access to multi-site resources, supporting advance reservations in a cluster has become increasingly important. In this paper we propose a new real-time divisible load scheduling algorithm that supports advance reservations in a cluster. The impact of advance reservations on system performance is systematically studied. Simulation results show that, with the proposed algorithm and appropriate advance reservations, the system performance could be maintained at the same level as the no reservation case. Thus, Our approach enforces the real-time agreement vis-a-vis addresses the under-utilization concerns.     

基于信任机制与遗传算法的渲染网格作业调度研究

信任关系是网格作业调度中一个很重要的因素,也是影响网格计算有效性和性能的关键技术之一。将信任机制引入到渲染网格作业调度中,建立渲染网格环境中基于信任机制的作业调度模型,在调度策略上对基本遗传算法进行了改进,提出了基于信任机制的遗传算法。实验结果表明,该算法可以提高任务完成率和平均信任效益,是适用于渲染网格的一种有效作业调度方法。     

计算网格中多编程环境的研究与实现

计算网格(也称元计算系统)聚集地理上分散的资源进行大型的分布式高性能计算。PVM和MPI是广泛使用的并行编程环境,它们需要作为并行计算的基本构建而集成到元计算系统中去。论文针对元计算资源的动态性、分布性、性能多变性和结点异构性等特点,实现了一个自适应的、一体化的多编程环境。论文论述了该多编程环境的体系结构,并利用代理技术实现远程编译、发现资源、屏蔽异构和优化调度。