A Launch-time Scheduling Heuristics for Parallel Applications on Wide Area Grids

Large and dynamic computational Grids, generally known as wide-area Grids, are characterized by a large availability, heterogene- ity on computational resources, and high vari- ability on their status during the time. Such Grid infrastructures require appropriate schedule mechanisms in order to satisfy the application performance requirements (QoS). In this paper we propose a launch-time heuristics to schedule component-based parallel applications on such kind of Grid. The goal of the proposed heuristics is threefold: to meet the minimal task computation- al requirement, to maximize the throughput between communicating tasks, and to evaluate on-the-fly the resource availability to minimize the aging effect on the resources state. We evaluate the proposed heuristics by simulations applying it to a suite of task graphs and Grid platforms randomly generated. Moreover, a further test was conducted to schedule a real application on a real Grid. Experimental results shown that the proposed solution can be a viable one.     

State-based predictions with self-correction on Enterprise Desktop Grid environments

The abundant computing resources in current organizations provide new opportunities for executing parallel scientific applications and using resources. The Enterprise Desktop Grid Computing (EDGC) paradigm addresses the potential for harvesting the idle computing resources of an organization’s desktop PCs to support the execution of the company’s large-scale applications. In these environments, the accuracy of response-time predictions is essential for effective metascheduling that maximizes resource usage without harming the performance of the parallel and local applications. However, this accuracy is a major challenge due to the heterogeneity and non-dedicated nature of EDGC resources. In this paper, two new prediction techniques are presented based on the state of resources. A thorough analysis by linear regression demonstrated that the proposed techniques capture the real behavior of the parallel applications better than other common techniques in the literature. Moreover, it is possible to reduce deviations with a proper modeling of prediction errors, and thus, a Self-adjustable Correction method (SAC) for detecting and correcting the prediction deviations was proposed with the ability to adapt to the changes in load conditions. An extensive evaluation in a real environment was conducted to validate the SAC method. The results show that the use of SAC increases the accuracy of response-time predictions by 35%. The cost of predictions with self-correction and its accuracy in a real environment was analyzed using a combination of the proposed techniques. The results demonstrate that the cost of predictions is negligible and the combined use of the prediction techniques is preferable.     

A Decentralized Computational Infrastructure for Grid-Based Parallel Asynchronous Iterative Applications

Parallel asynchronous iterative algorithms relax synchronization and communication requirements, and can potentially extend Desktop Grids beyond embarrassingly parallel applications to support a broader class of parallel iterative applications. This paper presents the design and implementation of CometG, a decentralized (peer-to-peer) computational infrastructure that extends Desktop Grid environments to support these applications. CometG provides a decentralized and scalable tuple space, efficient communication and coordination support, and application-level abstractions that can be used to implement Desktop Grid applications based on parallel asynchronous iterative algorithms using the master-worker/BOT paradigm. The deployment and evaluations of CometG and a CometG-based application in a wide-area environment using the PlanetLab [7] test bed, as well as a campus network are presented.     

工作站网络下三对角方程组并行求解

考虑工作站网络（ＮＯＷｓ）中三对角线性方程组的并行求解，基于最小秩解耦算法与分布治之并行计算模式，并行最小秩解耦算法（ＰＭＲＤ）。它在计算过程中保持原矩阵的结构特征，数值稳定性高，本文给出算法的数值特征分析以及计算与通讯复杂性分析并与Ｍｅｈｒｍａｎｎ分治算比较，所有算法由ＰＶＭ软件系统实现并在工作站网络中测试。     

FlexRay车载网络管理数据调度的研究

为确保汽车通信构架FlexRay的高效实时运行,需要尽可能地简化网络结构设计、降低通信复杂程度,良好的车载网络管理能够实现这些功能.简述了FlexRay通信协议和数据帧结构,在对比两种网络管理机制之后,提出了一种基于新规范的FlexRay网络管理方法.以其基本规范为依据,对网络管理数据的调度进行了分类,设计了数据的动态...     

An effective approximation algorithm for the Malleable Parallel Task Scheduling problem

The Malleable Parallel Task Scheduling problem (MPTS) is an extension of one of the most classic scheduling problems (P∥C_max). The only difference is that for MPTS, each task can be processed simultaneously by more than one processor. Such flexibility could dramatically reduce the makespan, but greatly increase the difficulty for solving the problem. By carefully analyzing some existing algorithms for MPTS, we find each of them suitable for some specific cases, but none is effective enough for all cases. Based on such observations, we introduce some optimization algorithms and improving techniques for MPTS, with their performance analyzed in theory. Combining these optimization algorithms and improving techniques gives rise to our novel scheduling algorithm OCM (Optimizations Combined for MPTS), a 2-approximation algorithm for MPTS. Extensive simulations on random datasets and SPLASH-2 benchmark reveal that for all cases, schedules produced by OCM have smaller makespans, compared with other existing algorithms.     

机群系统中有状态应用的调度算法研究

有状态的应用是当今因特网中较为典型的一种应用，利用机群系统可以有效地实现该种应用的扩容。请求调度算法是该种系统的关键技术，它决定机群系统的性能能否充分发挥。本文研究了轮循算法在有状态应用的使用。通过理论分析和实验仿真，指出了影响系统性能的主要因素。研究表明，该法在不同的应用条件下，有显著不同的性能表现。     

一个模拟网格上并行作业调度的离散事件系统

为评估计算网格上并行作业的无中心式调度,设计了基于离散事件的模拟系统J3S.离散事件包括作业提交、开始和终止,作业状态可以是等待、运行或完成.事件引起作业状态的变化 ,并触发网格层次或本地层次调度.作业由网格工作负荷模型产生,该模型基于并行计算机工作负荷模型构建.网格层次调度通过网格调度器之间的协作完成.本地层次调度模拟改进的装填法,使作业可有不同的调度优先级.网格资源性能可有差异,作业执行性能依赖作业指派.模块化实现使J3S可模拟网格上并行作业的无中心式调度的各种场景.     

云数据中心基于负载权重的负载均衡调度算法

针对现有云数据中心的多维资源利用不均衡问题,提出基于资源负载权重的动态多资源负载均衡调度算法。算法结合服务器各维度资源动态负载情况,构造层次分析法（AHP）判断矩阵来处理多维资源对于负载均衡影响权重大小,在此基础上综合考虑任务资源需求,将任务放置到合适服务器来改善资源利用,实现资源间负载均衡。平台仿真显示新算法可有效提高利用率低的资源的利用效率,在提高整体资源利用率、降低资源间负载不均衡率方面有优势。     

Performance Analysis of Round Robin Scheduling for Session-based Applications on Clusters

Nowadays session-based applications are one of the typical applications in the Internet,and people build such applications on clusters on concern of scalability. Scheduling in such a cluster is a key technology since system performance depends on it. In this paper,we investigate the Round-Robin algorithm in the context of Session-based applications. An analyzing model for such sys-tems is proposed. Through both theoretical analysis and simulation,we find the main factor for system performance. And the result also shows that this algorithm shows up with significantly different performance under various conditions.     

Approximation Schemes for Scheduling on Uniformly Related and Identical Parallel Machines

We give a polynomial approximation scheme for the problem of scheduling on uniformly related parallel machines for a large class of objective functions that depend only on the machine completion times, including minimizing the l_p norm of the vector of completion times. This generalizes and simplifies many previous results in this area.     

Impact of user patience on auto-scaling resource capacity for cloud services

An important feature of most cloud computing solutions is auto-scaling, an operation that enables dynamic changes on resource capacity. Auto-scaling algorithms generally take into account aspects such as system load and response time to determine when and by how much a resource pool capacity should be extended or shrunk. In this article, we propose a scheduling algorithm and auto-scaling triggering strategies that explore user patience, a metric that estimates the perception end-users have from the Quality of Service (QoS) delivered by a service provider based on the ratio between expected and actual response times for each request. The proposed strategies help reduce costs with resource allocation while maintaining perceived QoS at adequate levels. Results show reductions on resource-hour consumption by up to approximately 9% compared to traditional approaches.     

A Robust Compile Time Method for Scheduling Task Parallelism on Distributed Memory Machines

The problem of compile time scheduling of tasks of a program represented as a directed acyclic graph (DAG) is NP-hard in its general form. A number of approaches have been proposed which attempt to solve the problem either sub-optimally for general cases or optimally for restrictive special cases. But all the compile time approaches suffer due to the inability to accurately model the computation and communication costs of the target architecture. A desirable property of a compile time scheduling algorithm is robustness against the variations in the computation and communication costs so that the run time performance is close to the compile time estimates; this aspect of scheduling has been left open in the literature.This paper first introduces a compile time scheduling algorithm for a variable number of available processors and then examines the impact of change of computation and communication costs on the generated schedule. The cost variations for all the nodes and all the edges are assumed to be uniform (in other words, all the node costs change by the same ratio and the edge costs change by the same ratio). This sort of variations could occur due to the inaccuracies in estimating the instruction execution times or the message passing delays. The ratio of the schedule length obtained by the new schedule based on the modified costs to the schedule length obtained by using the modified costs on the original schedule (obtained by initial costs) is used as a measure of the robustness of the algorithm. The essential conditions for robustness of the proposed algorithm are discussed and are demonstrated through an experimental study.     

Analysis and Provision of QoS for Distributed Grid Applications

Grid computing provides the infrastructure necessary to access and use distributed resources as part of virtual organizations. When used in this way, Grid computing makes it possible for users to participate in collaborative and distributed applications such as tele-immersion, visualization, and computational simulation. Some of these applications operate in a collaborative mode, requiring data to be stored and delivered in a timely manner. This class of applications must adhere to stringent real-time constraints and Quality-of-Service (QoS) requirements. A QoS management approach is therefore required to orchestrate and guarantee the timely interaction between such applications and services. We discuss the design and a prototype implementation of a QoS system, and demonstrate how we enable Grid applications to become QoS compliant. We validate this approach through a case study of an image processing task derived from a nanoscale structures application.     

基于有色Petri网和IPSO的并行测试系统任务调度研究

并行测试以减少测试时间和降低测试成本的强大优势成为下一代自动测试系统ATS发展的热点;针对ATS中并行测试任务调度复杂、难以优化问题,提出了一种有色Petri网和改进粒子群优化(IPSO)算法相结合的任务调度优化算法;采用有色Petri网建立并行测试系统模型,得到并行测试的动态特性;采用IPSO算法搜索最优的任务调度路径,得到以测试时间最短为目标的最优任务调度方案;最后,将该算法应用到某型雷达电路板并行测试系统中,研究结果表明,与遗传算法GA相比,该算法效率更高,更利于工程应用。     

基于网络带宽约束的网格任务调度算法研究

在对网格计算任务调度算法的研究中,分析一般适应性调度启发算法即最小一最小(Min-min)算法的不足,在最小一最小(Min-min)算法的基础上,加入考虑QoS方面主要是网络带宽因素的任务调度算法.对以上算法做了实验并指出下一步的研究方向.     

Simultaneous Scheduling of Replication and Computation for Data-Intensive Applications on the Grid

Managing large datasets has become one major application of Grids. Life science applications usually manage large databases that should be replicated to scale applications. The growing number of users and the simple access to Internet-based application has stressed Grid middleware. Such environment are thus asked to manage data and schedule computation tasks at the same time. These two important operations have to be tightly coupled. This paper presents an algorithm (Scheduling and Replication Algorithm, SRA) that combines data management and scheduling using a steady-state approach. Using a model of the platform, the number of requests as well as their distribution, the number and size of databases, we define a linear program to satisfy all the constraints at every level of the platform in steady-state. The solution of this linear program will give us a placement for the databases on the servers as well as providing, for each kind of job, the server on which they should be executed. Our theoretical results are validated using simulation and logs from a large life science application. This work was supported in part by the ACI GRID and Grid5000 projects of the French Department of Research.     

Parallel communications for ATM network control and management

This paper describes an end-to-end parallel communications scheme based on a vector routing algorithm (VRA) for ATM network control and management. An information string is partitioned into m parts, which are then coded into k > m parts and sent out on k separate subchannels to the receiver. When m of the k parts are received correctly, the original information can be reconstructed. Two desirable effects are achieved in the context of ATM traffic control: (1) the burstiness of the source traffic can be smoothed out by the partition process; (2) the quality of service in terms of error, loss and delay can be controlled using the number of redundant routes k — m as a control parameter. Our results show that VRA is especially suitable for services with highly bursty traffic. We argue that several network management issues, including reliability, evolution and integration, security, and administration and billing can be addressed in a simple manner using the VRA framework.     

基于用户优先级的云计算任务调度策略

根据云计算环境中用户任务调度的不同需求,提出一种基于改进离散粒子群优化算法的任务调度策略,可实现在短时间内对云计算任务的相对较优调度。将用户费用与该任务的截止时间相结合,构建相对合理的用户优先级,以引导算法的适应度函数的偏好。引入重优化判断准则,在保证算法有能力跳出局部最优的同时保持解的多样性,最终求出满足用户优先级偏好的任务调度映射。仿真实验结果表明,该策略更符合云计算调度的复杂环境,能得到全局较优的任务调度方案。     

基于改进 NPGA 算法的多目标优化云任务调度算法

随着云计算的不断发展,传统的单目标优化下的任务调度已经不能满足用户的服务质量要求。论文选取运行时间、费用和负载均衡建立多目标优化的云任务调度模型,提出一种改进的多目标小生境 Pareto 遗传算法（NPGA）,采用相似任务序列交叉操作加快进化,再采用位移变异避免算法过早收敛。此外,通过自适应选取比较集合规模和小生境半径提高算法的收敛速度。仿真结果表明,改进后的 NPGA 算法在云调度中保持 Pareto 最优解的多样性和分布性更优。