Task granularity studies on a many-processor CRAY X-MP

A hybrid granularity model is proposed for general concurrent solution. It is applied to the triangular factorization of a dense matrix ranging in size from 4 to 1024. Concurrency is achieved at two levels: (1) with small (micro) task granularity and (2) with large (blocked) task granularity. Relevance to a many-processor CRAY X-MP is demonstrated by simulation.     

Using multiple CPUs for problem solving: Experiences in multitasking on the CRAY X-MP/48

The integration of vector computers into multiprocessor configurations allows the use of multiple high-speed processors in parallel for one program. There are two aspects which are considered to be important for an efficient use of multiprocessor configurations. First, the flexibility, speed and user friendliness of the available synchronization and communication primitives, and second, the user problems in detecting data dependencies and in translating programs correctly into the parallel form required by the system. This paper is intended to give an overview of our experiences in multitasking using up to four CPUs of a CRAY X-MP/48. The results gained by macrotasking and microtasking will be compared for program kernels and real-life application programs. Special attention is paid to the difficulties of using more than two CPUs in parallel.     

Multitasking: Experience with applications on a CRAY X-MP

On the multiprocessor vector-supercomputer CRAY X-MP, parallelism—beyond vectorization—can be exploited on the programming language level by two multitasking strategies: macrotasking and, more recently, microtasking. In this paper, multitasking results and experiences are presented which have been gained by applying these two implemented modes to linear-algebra and non-numerical algorithms as well as to a large fluid-flow simulation code. While comparing the concepts and realizations of macrotasking and microtasking, the features, tools, and problems of multitasking programming and the potential user benefit of these parallel processing techniques are discussed.     

测控系统的集群计算机的设计

并行程序开发的复杂性是并行系统得到广泛应用的主要障碍,文章针对靶场测控系统复杂多源信息的特点,提出了一种基于任务数据结构表并行计算;通过复杂多源信息的任务分解、任务数据结构的描述,并行计算模型设计以及负载平衡的实现等方面来实现机群系统并行计算;最后对系统进行了可扩展性和可用性分析(Availability),经分析系统能适应新任务的要求,具有很好的扩展性并具有连续不间断运行工作能力满足高可用性要求。     

Coupled-task scheduling on a single machine subject to a fixed-job-sequence

This paper investigates single-machine coupled-task scheduling where each job has two tasks separated by an exact delay. The objective of this study is to schedule the tasks to minimize the makespan subject to a given job sequence. We introduce several intriguing properties of the fixed-job-sequence problem under study. While the complexity status of the studied problem remains open, an O(n²) algorithm is proposed to construct a feasible schedule attaining the minimum makespan for a given permutation of 2n tasks abiding by the fixed-job-sequence constraint. We investigate several polynomially solvable cases of the fixed-job-sequence problem and present a complexity graph of the problem.     

软硬件划分中基于一种新的层次化控制数据流图的粒度变换

在一种控制数据流图定义的基础上,提出了面向软硬件划分中粒度选择的粒度变换方法．变换以层次化结点为中介,包括合并和展开操作,并遵循所指定的若干要求以保证变换前后的系统表示在形式和功能上的一致性．分析表明,这一变换过程简单快速,变换前后功能一致、形式统一,能很好地满足软硬件划分中粒度选择的需要．     

Implementing a simulation tool in a high-level language with no multitasking facilities

Ways to implement coroutines in a block-structured language with no multitasking facilities are presented. Coroutines are implemented as procedures. The reactivation points are kept in global variables, one variable for each procedure. Local variables whose values are required on re-entry are stored as STATIC objects. The variables or data of re-entrant coroutines are stored in an event list associated with each such coroutine. A procedure with several entries is a convenient mechanism to trap the primitive calls issued by the coroutines. This procedure returns to the master program by using a non-local GOTO. The implementation of the above in PL/I and C is described and a comparison is made with sequential Pascal. Ada includes constructs which satisfy most requirements.     

Gaussian process as complement to test functions for surrogate modeling

It is common for papers on surrogate fitting to select test functions for testing algorithms. This raises the issue of how well the algorithms generalize t     

Tradeoffs in granularity and parallelization for a Monte Carlo shower simulation code

The EGS4 code, developed at Stanford Linear Accelerator Center, simulates electron-photon cascading phenomena. The original code is inherently sequential: processing one particle at a time. This paper reports on a series of experiments in parallelizing different versions of EGS4. Our parallel experiments were run on a 30-processor Sequent Balance B21 and a 6-processor Symmetry S27. We have considered the following approaches for parallel execution of this application code:

1. (1) Original sequential version modified for parallel processing: 1 processor;

2. (2) Version 1 run multiprocessed: 1 to 29 processors;

3. (3) Sequential version modified for large-grain parallel processing: 1 procssor;

4. (4) Version 3 run using the Sequent Microtasking Library: 1 to 29 processors.

For each approach, we discuss the relative advantages and disadvantages in the areas of coding effort, understandability and portability, as well as performance, and outline a new parallelization approach we are currently pursuing based on Large-Grain Data Flow techniques.     

Use of concurrent and retrospective verbal protocols to investigate workers' thoughts during a manual-handling task

A manual handling task (an order picking task in a distribution warehouse) was investigated using concurrent verbal protocols and retrospective verbal protocols (applied at three different times after completion of the task). The study aimed to collect workers' thoughts during the task and to determine what workers are comfortable with reporting. Video recordings of the task were used as a means of investigating the accuracy of the information which was reported. The concurrent and retrospective verbal protocol methods were compared and it was found that concurrent reports could be used to collect a range of task related thoughts, although they only contained a proportion of the workers' thoughts during the task. Additional information could be obtained from retrospective reports, although these appeared to be vulnerable to bias and re-ordering of reported information. The retrospective reports produced whilst watching a video recording of the task were particularly vulnerable to bias from retrospective reasoning or the rationalisation of thoughts. Only one specific reference to posture or handling technique was made by any of the workers in their reports. There were also few references to load characteristics. Workers frequently reported aspects of the task that were easier to verbalise, such as the counting or labelling of items. It is not known whether the workers were only aware of these limited details of their tasks or whether they were only able to report limited details of the tasks because of factors affecting verbal reporting. This was an exploratory study which has collected valuable information from a small sample of experienced workers in relation to a common task from industry, but further study is needed to investigate the effects of factors such as load weight, load position and task duration on awareness and reporting of thoughts.     

Operator strategies in a dynamic environment in relation to an operator model

Results of an analysis of the cognitive activity of operators in controlling incidents in a Steam Generating Heavy Water Reactor in relation to a computer-based model of an operator (COSIMO, Cacciabue et al. 1992) are presented. The analysis focuses on information checking and actions to be taken in five incidents of varying levels of specification. Two experienced and two less-experienced operators participated in the experiments. Techniques of video and protocol analysis have been employed. Results indicate that operators use strategies in order to overcome the dynamic feature of incidents; in particular they break down problems into elementary problems, and they aim at maintaining a stable situation. A central characteristic is the importance of actions in terms of their timing and their results. Actions are carried out incrementally, and their effects then used as pointers to guide the course of future problem-solving and future actions. Problems may be postponed when they are not urgent, or when system latency does not allow an effect to be obtained following an action. Finally the paper discusses some elements that should be modified in order to increase the predictive capability of the COSIMO model.     

The influence of low frequency vibration on pilot performance (as measured in a fixed base simulator)

This report describes a study on the effects of low frequency vibration on pilot performance. Army pilots acted as subjects and flew simulated helicopter missions in a realistic fixed base simulator environment. While flying the two-hour missions pilots were exposed to vibration stimuli varying in frequency from 6 to 12 Hz and in amplitude from ±0·1 to ±0·3g (measured at the floor). Measurements were taken of the vibration tramsmissibility of the pilot's seat so that the vibration actually felt by the subject could be determined.

Missions involved transporting external loads in a simulated logistics environment. Performance was evaluated by measuring flight path deviations from prescribed en route, approach, and hover parameters.

The vibration stimuli used did not degrade performance. In fact, performance tended to improve with increased stress. It is hypothesized that this trend was due to motivation, i.e. as subjects felt the onset of fatigue they compensated by working harder and thus tended to improve their performance.

On about 6% of their scores pilots exhibited sudden lapses in their ability to respond to display indications. This resulted in poor scores in the midst of otherwise normal data. These lapses are probably of very short duration (seconds in length) and seem to occur randomly. This same effect was also observed in a previous study by the author. It may be possible that lapses of this type are related to so called ‘pilot error’ accidents. Further research is needed to verify this.     

From smartphones to iPad: Power users’ disposition toward mobile media devices

The spread of mobile media devices (MMDs), such as smartphones and tablet computers, has been markedly rapid around the world in recent years. The diffusion of mobile media technology has turned some early adopters into “power users,” using MMD more innovatively, efficiently and thoroughly than others. Going beyond the research of innovative users and heavy users, this study proposes the concept of power users by investigating the interactions between the power use of MMD and personality attributes, online activities as well as traditional media use. The findings showed that personality attributes were strong predictors for power use, which was also partially predicated by media multitasking and traditional media use. Research in this stream should assist in understanding the diffusion of mobile media technology and its impact on human interaction.     

禁止任务抢占法在实时嵌入式系统中的实现

提出了一种基于任务优先级的禁止抢占法,并分析了该方法的安全性和高效性,给出了具体实现的算法,在实际应用中有非常高的参考价值。     

To Schedule or to Execute: Decision Support and Performance Implications

This paper addresses a fundamental trade-off in dynamic scheduling between the cost of scheduling and the quality of the resulting schedules. The time allocated to scheduling must be controlled explicitly, in order to obtain good-quality schedules in reasonable times. As task constraints are relaxed, the algorithms proposed in this paper increase scheduling complexity to optimize longer and obtain high-quality schedules. When task constraints are tightened, the algorithms adjust scheduling complexity to reduce the adverse effect of long scheduling times on the schedule quality. We show that taking into account the scheduling time is crucial for honoring the deadlines of scheduled tasks. We investigate the performance of our algorithms in two scheduling models: one that allows idle-time intervals to exist in the schedule and another that does not. The model with idle-time intervals has important implications for dynamic scheduling which are discussed in the paper. Experimental evaluation of the proposed algorithms shows that our algorithms outperform other candidate algorithms in several parameter configurations.     

Intermediate-level vision tasks on a memory array architecture

With the fast advances in the area of computer vision and robotics there is a growing need for machines that can understand images at very high speed. A conventional von Neumann computer is not suitable for this purpose, because it takes a tremendous amount of time to solve most typical image analysis problems. Thus, it is now imperative to study computer vision in a parallel processing framework in order to reduce the processing time. In this paper we demonstrate the applicability of a simple memory array architecture to some intermediate-level computer vision tasks. This architecture, called theAccess Constrained Memory Array Architecture (ACMAA) has a linear array of processors which concurrently access distinct rows or columns of an array of memory modules. Because of its efficient local and global communication capabilities ACMAA is well suited for low-level as well as intermediate-level vision tasks. This paper presents algorithms for connected component labeling, determination of area, perimeter and moments of a labeled region, convex hull of a region, and Hough transform of an image. ACMAA is well suited to an efficient hardware implementation because it has a modular structure, simple interconnect and limited global control.     

Scheduling tasks of a parallel program in two-processor systems with use of cellular automata

In this paper, a cellular automaton (CA) is proposed as a tool for designing distributed scheduling algorithms for allocating parallel program tasks in multiprocessor systems. For this purpose, a program graph is considered as a CA containing elementary automata interacting locally according to some rules. In the first phase of the algorithm, effective rules for the CA are discovered by a genetic algorithm. In the second phase, the CA works as a distributed scheduler. In this phase, for any initial allocation of tasks in a multiprocessor system, the CA-based scheduler finds an allocation minimizing the total execution time of the program in a given system topology. The effectiveness of the proposed scheduling algorithm is shown for a number of program graphs scheduled in a two-processor system.     

利用批处理和VBS实现目录定时清理

批处理以其易用、功能丰富的特点,在计算机日常管理中发挥着重要作用。本文介绍了利用批处理命令和VBS实现按目录定时清理数据的方法。先利用DOS命令和VBS编写处理代码,形成批处理程序,然后通过Windows计划任务定时自动清理数据。实践表明,该方法有效提高了工作效率,减少了维护人员的工作量。