Heterogeneous distributed computing environments are well suited to meet the fast increasing computational demands. Task scheduling is very important for a heterogeneous distributed system to satisfy the large computational demands of applications. The performance of a scheduler in a heterogeneous distributed system normally has something to do with the dynamic task flow, that is, the scheduler always suffers from the heterogeneity of task sizes and the variety of task arrivals. From the long-term viewpoint it is necessary and possible to improve the performance of the scheduler serving the dynamic task flow. In this paper we propose a task scheduling method including a scheduling strategy which adapts to the dynamic task flow and a genetic algorithm which can achieve the short completion time of a batch of tasks. The strategy and the genetic algorithm work with each other to enhance the scheduler's efficiency and performance. We simulated a task flow with enough tasks, the scheduler with our strategy and algorithm, and the schedulers with other strategies and algorithms. We also simulated a complex scenario including the variant arrival rate of tasks and the heterogeneous computational nodes. The simulation results show that our scheduler achieves much better scheduling results than the others, in terms of the average waiting time, the average response time, and the finish time of all tasks.

In human-computer interaction, Fitts' law has been applied in one-dimensional pointing task evaluation for some decades, and the usage of effective target width (We) in Fitts' law has been accepted as an international standard in ISO standards 9241-9 [4]. However, the discussion on the concrete methods for calculating We has not been developed comprehensively nor have the different methods of calculation been integrated. Therefore, this paper focuses on a detailed description and a comparison of the two main We calculation methods. One method is mapping all the abscissa data in one united relative coordinate system to perform the calculation (called CC method) and the other is dividing the data into two groups and mapping them in two separate coordinate systems (called SC method). We tested the accuracy of each method and compared both methods in a highly controlled experiment. The experiments' results and data analysis show that the CC method is better than the SC method for human computer interface modeling. These results will be instrumental for future application of Fitts' law.

For execution of computation-intensive applications, one of the most important paradigms is to divide the application into a large number of small independent tasks and execute them on heterogeneous parallel computing environments (abbreviated by HPCEs). In this paper, we aim to execute independent tasks efficiently on HPCEs. We consider the problem to find a schedule that maximizes the throughput of task execution for a huge number of independent tasks. First, for HPCEs where the network forms a directed acyclic graph, we show that we can find, in polynomial time, a schedule that attains the optimal throughput. Secondly, for arbitrary HPCEs, we propose an (+ε)-approximation algorithm for any constant ε(ε>0). In addition, we also show that the framework of our approximation algorithm can be applied to other collective communications such as the gather operation.

To make the best use of the resources in a shared grid environment, an application scheduler must make a prediction of available performance on each resource. In this paper, we examine the problem of predicting available CPU performance in time-shared grid system. We present and evaluate a new and innovative method to predict the one-step-ahead CPU load in a grid. Our prediction strategy forecasts the future CPU load based on the variety tendency in several past steps and in previous similar patterns, and uses a polynomial fitting method. Our experimental results on large load traces collected from four different kinds of machines demonstrate that this new prediction strategy achieves average prediction errors which are between 22% and 86% less than those incurred by four previous methods.

Viable techniques such as dynamic voltage scaling (DVS) provide a new design technique to balance system performance and energy saving. In this paper, we extend previous works on task assignment problems for a set of linear-pipeline tasks over a set of processors. Different from previous works, we revisit the problems with two additional system factors: deadline and energy-consumption, which are key factors in real-time and power-aware computation. We propose an O(nm2) time complexity algorithm to determine optimal task-assignment and speed-setting schemes leading to minimal energy consumption, for a given set of m real-time tasks running on n identical processors (with or without DVS supports). The same result can be extended to a restricted form of heterogeneous processor model. Meanwhile, we show that on homogeneous processor model more efficient algorithms can be applied and result in time complexity of O(m2) when m ≤ n. For completeness, we also discuss cases without contiguity constraints. We show under such cases the problems become at least as hard as NP-hard.

A hard real-time system is one whose correctness depends not only on the logical result, but also when the results are produced. While many techniques have been proposed for single processor real-time systems, multiprocessor systems have not been studied so extensively. In this paper, we mainly propose two variant (DCTS) by using the Early-Release-Fair (ERfair) and Proportionate-fair (Pfair) model with integral assumptions for identical multi-processor real-time systems. ERfair is a scheduling model for real-time tasks on a multiprocessor system. On the different definitions of distance constraint, we propose two efficient scheduling algorithms designed to probe whether the distance constraints of all ER-fair tasks can be guaranteed. If the distance constraints cannot be guaranteed, then the proposed algorithms gather the unfeasible tasks and inflate them with a reweighting function. The proposed algorithms are linear-time and most suitable for dynamic systems. The experimental results reveal that the proposed algorithms increase significantly the ratio of schedulable task sets.

Designable task allocation systems which consist of identical agents using stochastic automata are suggested. From the viewpoint of the group response and the individual behavior, the performances of a simple model and an improved one are compared numerically. Robots experiments are performed to compare between the two models.

To obtain a robust acoustic model for a certain speech recognition task, a large amount of speech data is necessary. However, the preparation of speech data including recording and transcription is very costly and time-consuming. Although there are attempts to build generic acoustic models which are portable among different applications, speech recognition performance is typically task-dependent. This paper introduces a method for automatically building task-dependent acoustic models based on selective training. Instead of setting up a new database, only a small amount of task-specific development data needs to be collected. Based on the likelihood of the target model parameters given this development data, utterances which are acoustically close to the development data are selected from existing speech data resources. Since there are too many possibilities for selecting a data subset from a larger database in general, a heuristic has to be employed. The proposed algorithm deletes single utterances temporarily or alternates between successive deletion and addition of multiple utterances. In order to make selective training computationally practical, model retraining and likelihood calculation need to be fast. It is shown, that the model likelihood can be calculated fast and easily based on sufficient statistics without the need for explicit reconstruction of model parameters. The algorithm is applied to obtain an infant- and elderly-dependent acoustic model with only very few development data available. There is an improvement in word accuracy of up to 9% in comparison to conventional EM training without selection. Furthermore, the approach was also better than MLLR and MAP adaptation with the development data.

Recent improvements in the performance of end-computers and networks have made it feasible to construct a grid system over the Internet. A grid environment consists of many computers, each having a set of components and a distinct performance. These computers are shared among many users and managed in a distributed manner. Thus, it is important to focus on a situation in which the computers are used unevenly due to decentralized management by different task schedulers. In this study, which is a preliminary investigation of the performance of task allocation schemes employed in a decentralized environment, the average execution time of a long-lived task is analytically derived using the M/G/1-PS queue. Furthermore, assuming a more realistic condition, we evaluate the performance of some task allocation schemes adopted in the analysis, and clarify which scheme is applicable to a realistic grid environment.

Efficient task scheduling is critical for achieving high performance in grid computing systems. Existing task scheduling algorithms for grid environments usually assume that the performance prediction for both tasks and resources is perfectly accurate. In practice, however, it is very difficult to achieve such an accurate prediction in a heterogeneous and dynamic grid environment. Therefore, the performance of a task scheduling algorithm may be significantly influenced by prediction inaccuracy. In this paper, we study the influence of inaccurate predictions on task scheduling in the contexts of task selection and processor selection, which are two critical phases in task scheduling algorithms. We develop formulas for the misprediction degree, which is defined as the probability that the predicted values for the performances of tasks and processors reveal different orders from their real values. Based on these formulas, we also investigate the effect of several key parameters on the misprediction degree. Finally, we conduct extensive simulation for the sensitivities of some existing task scheduling algorithms to the prediction errors.

The technological development of wireless environment has made real-time multimedia communications possible in wireless networks. Many studies have been done on real-time communications in wireless networks in order to overcome a higher bit error rate in wireless channels. However, none of work deals with firm real-time communications which can be applied to multimedia communications. In this paper, we propose an adaptive error correcting scheme for firm real-time multimedia communications in wireless networks in order to maximize the expected net profit. The proposed scheme adaptively selects an error correcting code under the current air state and the message state of a message stream. Throughout simulation results, we show that the suggested scheme provides more profit than single error-correcting code schemes.

This paper presents a distributed task assignment algorithm in a logical unidirectional ring, which guarantees that almost all tasks are assigned to servers with the first come first served (FCFS) policy without a global clock. A task assignment for a process is obtained in the time period needed for a message to circle the ring. This time period is almost optimal for a unidirectional ring. The FCFS policy is very important in terms of task fairness and can also avoid starvation and provide an efficient response time. Simulation results show that the algorithm generally works better than conventional task assignment or load balancing schemes with respect to both mean response time and task fairness.

This paper proposes a new framework to produce humanoid animations for simulating human tasks. Natural working movements are generated via management of motion capture data with our simulation package. An extensible middleware controls reactive human behaviors, and all processes of simulation in a cyber factory are controlled through XML documents including motions, scene objects, and behaviors. This package displays simulation using Web3D technology and X3D specifications which can supply a common interface for customizing cyberworlds.

The supertask approach was proposed by Moir and Ramamthy as a means of supporting non-migratory tasks in Pfair-scheduled systems. In this approach, tasks bound to the same processor are combined into a single server task, called a supertask, which is scheduled as an ordinary Pfair task. When a supertask is scheduled, one of its component tasks is selected for execution. In previous work, Holman et al. showed that component-task deadlines can be guaranteed by inflating each supertask's utilization. In addition, their experimental results showed that the required inflation factors should be small in practice. Consequently, the average inflation produced by their rules is much greater than that actually required by the supertasks. In this paper, we first propose a notion of Transient Behavior Prediction for supertasks, which predicts the latest possible finish time of subtasks that belong to supertasks. On the basis of the notion, we present an efficient schedulability algorithm for Pfair supertasks in which the deadlines of all component tasks can be guaranteed. In addition, we propose a task merging process which combines the unschedulable supertasks with some Pfair tasks; hence, a newly supertask can be scheduled in the system. Finally, we propose the new reweighting functions that can be used when the previous two methods fail. Our reweighting functions produce smaller inflation factor than the previous work does. To demonstrate the efficacy of the supertasking approach, we present the experimental evaluations of our algorithm, which decreases substantially a number of reweights and the size of inflation when there are many supertasks in the Pfair-scheduled systems.

Global computing system (GCS) harnesses the idle CPU resources of clients connected to Internet for solving large problems that require high volume of computing power. Since GCS scale to millions of clients, many projects usually adopt coarse-grained scheduling in order to reduce server-side contention at the expense of sacrificing the degree of parallelism and wasting CPU resources. In this paper, we propose a new type of client, i.e., a scheduling proxy that enables adaptive-grained scheduling between the server and clients. While the server allocates coarse-grained work units to scheduling proxies alone, clients download fine-grained work units from a relatively nearby scheduling proxy not from the distant server. By computation of small work units at client side, the turnaround time of work unit can be reduced and the waste of CPU time by timeout can be minimized without increasing the performance cost of contention at the server. In addition, in order not to lose results in the failure of scheduling proxies, we suggest a technique of result caching in clients.

This paper presents a prototype of a distributed 3D rendering system in a hierarchical Grid environment. 3D rendering with massive data sets is a computationally intensive task. In order to make full use of computational resources on Grids, a hierarchical system architecture is designed to run over multiple clusters. This architecture involves both sort-first and sort-last parallel rendering algorithms to achieve excellent scalability, rendering performance and load balance.

In a Distributed Computing Systems (DCS) tasks submitted to it, are usually partitioned into different modules and these modules may be allocated to different processing nodes so as to achieve minimum turn around time of the tasks utilizing the maximum resources of the existing system such as CPU speed, memory capacities etc. The problem lies on how to obtain the optimal allocation of these multiple tasks by keeping in mind that no processing node is overloaded due to this allocation. This paper proposes an algorithm A*RS, using well-known A*, which aims to reduce the search space and time for task allocation. It aims at minimization of turn around time of tasks in the way so that processing nodes do not become overloaded due to this allocation. Our experimental results justify the claims with necessary supports by comparing it with the earlier algorithm for multiple tasks allocation.

This paper presents a new submesh allocation scheme for mesh connected multicomputer systems, called CFSL-TR (Classified Free Submesh List-Task Relocation), which reduces task waiting time in two aspects, shortening submesh search time and reducing the submesh allocation delay caused by external fragmentation. This scheme classifies independent free submeshes by their types: square, horizontal rectangle, or vertical rectangle. Then it searches for the best-fit submesh only from one list depending on the type of the given task, thus saving submesh searching time. If no suitable submeshes are found, it is most likely caused by external fragmentation. In such a case, our scheme relocates the tasks being executed to free submeshes and combines the newly available submesh with other fragmented ones to form a larger submesh. This allows allocation of the task, otherwise to be put on the queue, hence reducing the submesh allocation delay. Through simulation, we show that our scheme helps reduce task waiting time and that it is by far more effective to reduce the submesh allocation delay caused by external fragmentation rather than to reduce submesh search time for reduction of the task waiting time.

In recent years, there has been a rapid and widespread proliferation of non-traditional embedded computing platforms such as digital camcorders, cellular phones, and portable medical devices. As applications become increasingly sophisticated and processing power increases, the application designer has to rely on the services provided by the real-time operating systems (RTOSs). These RTOSs must not only provide predictable services but must also be efficient and small in size. Kernel services should also be deterministic by specifying how long each service call will take to execute. Having this information allows the application designers to better plan their real-time application software so as not to miss the deadline of each task. In this paper, we propose a generalized deterministic scheduling algorithm that makes the task scheduling time constant irrespective of the number of tasks created in an application. The proposed algorithm eliminates the restriction on the maximum number of task priorities imposed on the existing ones, without additional memory overhead.

The paper develops the transformation rules in order to use the Stochastic Petri Net model to evaluate the performance of various task scheduling algorithms. The transformation rules are applied to DFRN scheduling algorithm to investigate its effectiveness. The performance comparison reveals that the proposed approach provides very accurate evaluation for the scheduling algorithm when the Communication to Computation Ratio value is small.