Analysis of fork-join program response times on multiprocessors

Models for two processor sharing policies called task scheduling processor sharing and job scheduling processor sharing are developed and analyzed. The first policy schedules each task independently and allows parallel execution of an individual program, whereas the second policy schedules each job as a unit, thereby not allowing parallel execution of an individual program. It is found that task scheduling performs better than job scheduling for most system parameter values. The performance of the task scheduling processor sharing is compared to a first come first serve policy. First come first serve performs better than processor sharing over a wide range of system parameters. Processor sharing performs best when the task service time variability is high. The performance of processor sharing and first come first serve is studied with two classes of jobs, and for when a specific number of processors is statically assigned to each of the classes     

Adaptive hierarchical scheduling policy for enterprise grid computing systems

In an enterprise grid computing environments, users have access to multiple resources that may be distributed geographically. Thus, resource allocation and scheduling is a fundamental issue in achieving high performance on enterprise grid computing. Most of current job scheduling systems for enterprise grid computing provide batch queuing support and focused solely on the allocation of processors to jobs. However, since I/O is also a critical resource for many jobs, the allocation of processor and I/O resources must be coordinated to allow the system to operate most effectively. To this end, we present a hierarchical scheduling policy paying special attention to I/O and service-demands of parallel jobs in homogeneous and heterogeneous systems with background workload. The performance of the proposed scheduling policy is studied under various system and workload parameters through simulation. We also compare performance of the proposed policy with a static space–time sharing policy. The results show that the proposed policy performs substantially better than the static space–time sharing policy.     

Measurement,modeling and enhancement of BitTorrent-based VoD system

BitTorrent is one of the most popular Peer-to-Peer (P2P) applications for file sharing over the Internet. Video files take up a large proportion of space among the BitTorrent shared files. Recently, BitTorrent has attracted researchers’ interests, as an alternative method of providing video on demand (VoD) service. In this paper, we concentrate on enabling BitTorrent to support VoD service in existing swarms while maintaining the download efficiency of file-sharing users. We first examine the content properties of the BitTorrent system to explore the demands and challenges of VoD service in BitTorrent swarms. The efficiency of BitTorrent for various piece selection policies is then compared through measurement on PlanetLab. We also use an optimization mathematical model to analyze the hybrid BitTorrent system in which downloading peers and streaming peers coexist. Both measurement results and model analysis indicate the problem of system efficiency decline in the BitTorrent-based VoD systems, in comparison with the original BitTorrent file-sharing system. Our proposed approach, unlike existing strategies that are limited to changing the piece selection policy to allow BitTorrent to support streaming services, modifies both piece and peer selection policies to provide a “streaming while downloading” service in the BitTorrent system with downloading peers. For the peer selection policy, a CAP (Closest-Ahead Peers) method is applied to make better use of the peers’ upload bandwidths. For the piece selection policy, a sliding window-based hybrid method that combines the rarest-first policy with the sequential policy is proposed. To demonstrate the performance of our proposed approach, an evaluation is made using various metrics on PlanetLab. The results show that our proposed method has higher throughput and better streaming continuity than the sequential policy and BiToS.     

Extending goal-oriented parallel computer job scheduling policies to heterogeneous systems

Several existing parallel computer systems partition their system resources or consist of systems from different geographical locations. This work is focusing on extending the original goal-oriented parallel computer job scheduling policies to cover such systems. The goal-oriented parallel computer job scheduling policies are proposed recently to handle conflicting objectives by utilizing a combinatorial search technique to find the most compromise schedule within a time limit. In this paper, some modifications to the original goal-oriented parallel computer job scheduling policy design are proposed and evaluated. The proposed policy is evaluated against basic priority backfilling techniques widely used in the field. Both homogeneous and heterogeneous parallel computer systems in terms of the computing power are evaluated. And, the design decision on the partition selection heuristic is also evaluated in this study. The experimental results show that the proposed policy produces good scheduling performances even when inaccurate runtime information is used.     

A hybrid particle swarm optimization for job shop scheduling problem

A hybrid particle swarm optimization (PSO) for the job shop problem (JSP) is proposed in this paper. In previous research, PSO particles search solutions in a continuous solution space. Since the solution space of the JSP is discrete, we modified the particle position representation, particle movement, and particle velocity to better suit PSO for the JSP. We modified the particle position based on preference list-based representation, particle movement based on swap operator, and particle velocity based on the tabu list concept in our algorithm. Giffler and Thompson’s heuristic is used to decode a particle position into a schedule. Furthermore, we applied tabu search to improve the solution quality. The computational results show that the modified PSO performs better than the original design, and that the hybrid PSO is better than other traditional metaheuristics.     

Processor Allocation in Multiprogrammed Distributed-Memory Parallel Computer Systems

In this paper, we examine three general classes of space-sharing scheduling policies under a workload representative of large-scale scientific computing. These policies differ in the way processors are partitioned among the jobs as well as in the way jobs are prioritized for execution on the partitions. We consider new static, adaptive and dynamic policies that differ from previously proposed policies by exploiting user-supplied information about the resource requirements of submitted jobs. We examine the performance characteristics of these policies from both the system and user perspectives. Our results demonstrate that existing static schemes do not perform well under varying workloads, and that the system scheduling policy for such workloads must distinguish between jobs with large differences in execution times. We show that obtaining good performance under adaptive policies requires somea prioriknowledge of the job mix in these systems. We further show that a judiciously parameterized dynamic space-sharing policy can outperform adaptive policies from both the system and user perspectives.     

Using aggregation to construct periodic policies for routing jobs to?parallel servers with deterministic service times

The problem of routing deterministic arriving jobs to parallel servers with deterministic service times, when the job arrival rate equals the total service capacity, requires finding a periodic routing policy. Because there exist no efficient exact procedures to minimize the long-run average waiting time of arriving jobs, heuristics to construct periodic policies have been proposed. This paper presents an aggregation approach that combines servers with the same service rate, constructs a policy for the aggregated system, and then disaggregates this policy into a feasible policy for the original system. Computational experiments show that using aggregation not only reduces average waiting time but also reduces computational effort.     

A trace-driven simulation study of dynamic load balancing

A trace-driven simulation study of dynamic load balancing in homogeneous distributed systems supporting broadcasting is presented. Information about job CPU and input/output (I/O) demands collected from production systems is used as input to a simulation model that includes a representative CPU scheduling policy and considers the message exchange and job transfer cost explicitly. Seven load-balancing algorithms are simulated and their performances compared. Load balancing is capable of significantly reducing the mean and standard deviation of job response times, especially under heavy load, and for jobs with high resource demands. Algorithms based on periodic or nonperiodic load information exchange provide similar performance, and, among the periodic policies, the algorithms that use a distinguished agent to collect and distribute load information cut down the overhead and scale better. With initial job placements only, source initiative algorithms were found to perform better than server initiative algorithms. The performances of all hosts, even those originally with light loads, are generally improved by load balancing     

Real-time integrated production-scheduling and maintenance-planning in a flexible job shop with machine deterioration and condition-based maintenance

The introduction of modern technologies in manufacturing is contributing to the emergence of smart (and data-driven) manufacturing systems, known as Industry 4.0. The benefits of adopting such technologies can be fully utilized by presenting optimization models in every step of the decision-making process. This includes the optimization of maintenance plans and production schedules, which are two essential aspects of any manufacturing process. In this paper, we consider the real-time joint optimization of maintenance planning and production scheduling in smart manufacturing systems. We have considered a flexible job shop production layout and addressed several issues that usually take place in practice. The addressed issues are: new job arrivals, unexpected due date changes, machine degradation, random breakdowns, minimal repairs, and condition-based maintenance (CBM). We have proposed a real-time optimization-based system that utilizes a modified hybrid genetic algorithm, an integrated proactive-reactive optimization model, and hybrid rescheduling policies. A set of modified benchmark problems is used to test the proposed system by comparing its performance to several other optimization algorithms and methods used in practice. The results show the superiority of the proposed system for solving the problem under study. The results also emphasize the importance of the quality of the generated baseline plans (i.e., initial integrated plans), the use of hybrid rescheduling policies, and the importance of rescheduling times (i.e., reaction times) for cost savings.     

Mesh网均等分区策略

在大规模并行计算机系统中，处理机资源可能被多个用户作业竞争，操作系统必须采用一种处理机分配策略确定多少和哪些处理机分配给一个作业。文中针对大规模、消息通信并行计算机提出了矩形和非矩形两种处理机分配策略，这两种策略均满足对每个用户所分配处理机数的公平性以及处理机分配的邻近性。     

校园网格环境下软件License调度算法及策略研究

应用软件License是昂贵的计算资源,校园网格是分布、共享的应用环境,在该环境下研究License的管理有着重要的意义。本文提出了支持License管理策略的校园网格管理系统体系结构,提出了基于角色和等待时间的网格作业动态优先级计算方法,提出了基于资源稀缺性和负载的License预留策略及剥夺策略,并在此基础上提出了基于动态优先级和集群负载的网格资源调度算法PLS。提出的算法和策略在校园网格进行了部署和应用,证明了其可以提高资源利用率,实现不同等级用户服务的差异性,保证同等级用户之间的公平性。     

The value of information sharing in a supply chain with a seasonal demand process

This paper considers a two echelon seasonal supply chain model that consists of one supplier and one retailer, with the assumption that external demand from the customer follows a seasonal autoregressive moving average (SARMA) process, including marketing actions that cannot be deduced from the other parameters of the demand process. In our model, the supplier and the retailer employ order-up-to policy to replenish their inventory. In order to evaluate the value of information sharing in a two echelon seasonal supply chain, three levels of information sharing proposed by Yu, Yan, and Cheng (2002) are used. The results for optimal inventory policies under these three levels of information sharing are derived. We show that the seasonal effect has an important impact on optimal inventory policies of the supplier under the three levels of information sharing. Our findings also demonstrate that the replenishment of lead time must be less than the seasonal period in order to benefit from information sharing. Thus, this result provides managers with managerial insights to improve supply chain performance through information sharing integration partnerships.     

Recovery analysis of data sharing systems under deferred dirty pagepropagation policies

In a multinode data sharing environment, different buffer coherency control schemes based on various lock retention mechanisms can be designed to exploit the concept of deferring the propagation or writing of dirty pages to disk to improve normal performance. Two types of deferred write policies are considered. One policy only propagates dirty pages to disk at the times when dirty pages are flushed out of the buffer under LRU buffer replacement. The other policy also performs writes at the times when dirty pages are transferred across nodes. The dirty page propagation policy can have significant implications on the database recovery time. In this paper, we provide an analytical modeling framework for the analysis of the recovery times under the two deferred write policies. We demonstrate how these policies can be mapped onto a unified analytic modeling framework. The main challenge in the analysis is to obtain the pending update count distribution which can be used to determine the average numbers of log records and data I/Os needed to be applied during recovery. The analysis goes beyond previous work on modeling buffer hit probability in a data sharing system where only the average buffer composition, not the distribution, needs to be estimated, and recovery analysis in a single node environment where the complexities on tracking the propagation of dirty pages across nodes and the buffer invalidation effect do not appear     

基于Objrmap的区分实时进程页的页面置换算法

本文深入分析了Linux 2.6的新技术--基于对象的逆向映射Objrmap的基本原理,探讨了区分实时进程页的置换策略的可能性,并提出了基于区分实时进程页的页面置换算法,增 强了Linux系统的实时性。改进的算法具有优异的时空性能,同时不影响系统的高并发度和进程间的内存共享。实验表明,改进算法在重载条件下,实时进程的缺页次数明显少于 原算法。     

Asymptotic convergence of scheduling policies with respect to slowdown

We explore the performance of an M/GI/1 queue under various scheduling policies from the perspective of a new metric: the slowdown experienced by the largest jobs. We consider scheduling policies that bias against large jobs, towards large jobs, and those that are fair, e.g., processor-sharing (PS). We prove that as job size increases to infinity, all work conserving policies converge almost surely with respect to this metric to no more than 1/(1−ρ), where ρ denotes the load. We also find that the expected slowdown under any work conserving policy can be made arbitrarily close to that under PS, for all job sizes that are sufficiently large.     

Evolutionary ensemble of diverse artificial neural networks using speciation

Recently, many researchers have designed neural network architectures with evolutionary algorithms but most of them have used only the fittest solution of the last generation. To better exploit information, an ensemble of individuals is a more promising choice because information that is derived from combining a set of classifiers might produce higher accuracy than merely using the information from the best classifier among them. One of the major factors for optimum accuracy is the diversity of the classifier set. In this paper, we present a method of generating diverse evolutionary neural networks through fitness sharing and then combining these networks by the behavior knowledge space method. Fitness sharing that shares resources if the distance between the individuals is smaller than the sharing radius is a representative speciation method, which produces diverse results than standard evolutionary algorithms that converge to only one solution. Especially, the proposed method calculates the distance between the individuals using average output, Pearson correlation and modified Kullback–Leibler entropy to enhance fitness sharing performance. In experiments with Australian credit card assessment, breast cancer, and diabetes in the UCI database, the proposed method performed better than not only the non-speciation method but also better than previously published methods.     

The processor working set and its use in scheduling multiprocessorsystems

The concept of a processor working set (PWS) as a single value parameter for characterizing the parallel program behavior is introduced. Through detailed experimental studies of different algorithms on a transputer-based multiprocessor machine, it is shown that the PWS is a robust measure for characterizing the workload of a multiprocessor system. It is shown that processor allocation strategies based on the PWS provide significantly better throughput-delay characteristics. The robustness of PWS is further demonstrated by showing that allocation policies that allocate processors more than the PWS are inferior in performance to those that never allocate more than the PWS-even at a moderately low load. Based on the results, a simple static allocation policy that allocates the PWS at low load and adaptively fragments at high load to one processor per job is proposed     

Job scheduling in mesh multicomputers

A new approach for dynamic job scheduling in mesh-connected multiprocessor systems, which supports a multiuser environment, is proposed in this paper. Our approach combines a submesh reservation policy with a priority-based scheduling policy to obtain high performance in terms of high throughput, high utilization, and low turn-around times for jobs. This high performance is achieved at the expense of scheduling jobs in a strictly fair, FCFS fashion; in fact, the algorithm is parameterized to allow trade-offs between performance and (short-term) POPS fairness. The proposed scheduler can be used with any submesh allocation policy. A fast and efficient implementation of the proposed scheduler has also been presented. The performance of the proposed scheme has been compared with the FCFS policy, the only existing scheduling strategy for meshes, to demonstrate the effectiveness of the proposed approach. Simulation results indicate that our scheduling strategy outperforms the FCFS policy significantly. Specifically, our strategy significantly reduces the average waiting delay of jobs over the FCFS policy. The fast implementation of the proposed scheduler results in low allocation and deallocation time overhead, as well as low space overhead     

Embedded simulation on a multiprocessor job scheduling system with inspection

This paper first develops architecture for a multiprocessor job scheduling system with an embedded simulation technique. The architecture provides a shell for applications that are characterized by two scheduling policies, a heuristic algorithm policy and a First-In-First-Out (FIFO) policy. These policies are implemented in the simulation model by using the embedded technique. The paper evaluates these two policies using the queue length, waiting time and flow time as the criteria to compare the performance of these two scheduling policies. Next we designed two simulation situations using two different real world applications. The purpose is to examine the performances of multiprocessor systems with and without inspection operations and two different scheduling policies. The two applications, berth allocation for the container terminal operations and production scheduling arrangement in an Original Equipment Manufacturer (OEM) power supply factory, are studied. The final results show that a proper scheduling policy will perform better than the traditional FIFO approach for a multiprocessor system. Our study also provides guidelines on balancing a system with the addition of a final inspection activity.     

Dynamic shortest path problems: Hybrid routing policies considering network disruptions

Traffic network disruptions lead to significant increases in transportation costs. We consider networks in which a number of links are vulnerable to these disruptions leading to a significantly higher travel time on these links. For these vulnerable links, we consider known link disruption probabilities and knowledge of transition probabilities for recovering from or getting into a disruption. We develop a framework based on dynamic programming in which we formulate and evaluate different known online and offline routing policies. Next to this, we develop computation-time-efficient hybrid routing policies. To test the efficiency of the different routing policies, we develop a test bed of networks based on a number of characteristics and analyze the results in terms of routes, cost performance and calculation times. Our results show that a significant part of the cost reduction can be obtained by considering only a limited part of the network in detail. The performance of our proposed hybrid policy is only slightly worse than the optimal policy.