• Proceedings of the IEEK Conference
• /
• 2002.07a
• /
• pp.176-179
• /
• 2002

In recent years, an increasing amount of computer network research has focused on the problem of cluster system in order to achieve higher performance and lower cost. The load unbalance is the major defect that reduces performance of a cluster system that uses parallel program in a form of SPMD (Single Program Multiple Data). Also, the load unbalance is a problem of MPP (Massive Parallel Processors), and distributed system. The cluster system is a loosely-coupled distributed system, therefore, it has higher communication overhead than MPP. Dynamic load balancing can solve the load unbalance problem of cluster system and reduce its communication cost. The cluster systems considered in this paper consist of P heterogeneous nodes connected by a switch-based network. The master node can predict the average execution time of tasks for each slave node based on the information from the corresponding slave node. Then, the master node redistributes remaining tasks to each node considering the predicted execution time and the communication overhead for task migration. The proposed dynamic load balancing uses execution time prediction to optimize the task redistribution. The various performance factors such as node number, task number, and communication cost are considered to improve the performance of cluster system. From the simulation results, we verified the effectiveness of the proposed dynamic load balancing algorithm.

• PNF and Movement
• /
• v.10 no.1
• /
• pp.25-33
• /
• 2012

Purpose : The purpose of this study was to compare the effect of treadmill training and cognitive task with in the course of treadmill training at the same time with chronic stroke patients. Methods : Fourteen chronic stroke patients participated. Participants were randomly assigned to the control and experimental group(7 experimental, 7 control). All of participants were in-patients at local hospital and had been receiving a traditional rehabilitation program, five days a week. The both groups have undergone 4weeks. The experimental group trained in treadmill and cognitive task at the same time, but control group trained only treadmill. 10m walking test, Timed Up & Go (TUG) test and 6 Minutes walking(6M walking) test to measure the walking speed, dynamic balance and waling endurance ability were carried out before and after the training. Results : The result of the study were as follow:10m walking test were significantly increased both groups(p<.01), but not significant between groups(p>.05). TUG test were significantly increased both groups(p<.001) and between groups(p<.01). 6M walking test were significantly increased both groups(p<.001), but not significant between groups(p>.05). Conclusion : Ahead of return to the community to patients with stroke, cognitive task with in the course of treadmill training at the same time was effective in improving the dynamic balance ability.

• International Journal of Contents
• /
• v.6 no.3
• /
• pp.19-24
• /
• 2010

Task scheduling is an integrated component of computing with the emergence of grid computing. In this paper, we address two different task scheduling models, which are static Round-Robin (RR) and dynamic Fastest Site First (FSF) task scheduling method, using extended timed marked graphs, which is a special case of Stochastic Petri Nets (SPN). Stochastic reward nets (SRN) is an extension of SPN and provides compact modeling facilities for system analysis. We build hierarchical SRN models to compare two task scheduling methods. The upper level model simulates task scheduling and the lower level model implements task serving process for different sites with multiple servers. We compare these two models and analyze their performances by giving reward measures in SRN.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.10
• /
• pp.930-939
• /
• 2004

In this paper, both dynamic constraints and kinematic constraints are considered for the analysis of manipulability of robotic systems comprised of multiple cooperating arms. Given bounds on the torques of each Joint actuator for every robot, the purpose of this study is to drive the bounds of task-space acceleration of object carried by the system. Bounds on each joint torque, described as a polytope, is transformed to the task-space acceleration through matrices related with robot dynamics, robot kinematics, object dynamics, grasp conditions, and contact conditions. A series of mathematical manipulations including the procedure calculating minimum infinite-norm solution of linear equation is applied to get the reachable acceleration bounds from given actuator dynamic constrains. Several examples including two robot systems as well as three robot system are shown with the assumptions of complete-constraint contact model(or' very soft contact') and insufficient or proper degree of freedom robot.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.6
• /
• pp.273-283
• /
• 2008

Many embedded real-time systems have adopted processors supported with dynamic voltage scaling(DVS) recently. Power is one of the important metrics for optimization in the design and operation of embedded real-time systems. We can save considerable energy by using slowdown of processor supported with DVS. In this paper, we propose heuristic algorithms to calculate task slowdown factors for an efficient energy consumption in embedded real-time systems with task synchronization. The previous algorithm has a following constraint : given the tasks are ordered in a nondecreasing order of their relative deadline, the task slowdown factors computed are in a nonincreasing order. In this paper, we relax the constraint and propose heuristic algorithms which have the same time complexity that previous algorithm has and can save more energy. Experimental results show that the proposed algorithms are energy efficient.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.5
• /
• pp.866-874
• /
• 2006

This paper proposes a new Dynamic Voltage Scaling(DVS) algorithm to achieve low-power scheduling of aperiodic hard real-time tasks. Aperiodic tasks schedulingcannot be applied to the conventional DVS algorithm and result in consuming energy more than periodic tasks because they have no period, non predictable worst case execution time, and release time. In this paper, we defined Virtual Periodic Task Set(VTS) which has constant period and worst case execution time, and released aperiodic tasks are assigned to this VTS. The period and worst case execution time of the virtual task can be obtained by calculating task utilization rate of both periodic and aperiodic tasks. The proposed DVS algorithm scales the frequency of both periodic and aperiodic tasks in VTS. Simulation results show that the energy consumption of the proposed algorithm is reduced by 11% over the conventional DVS algorithm for only periodic task.

• Proceedings of the KAIS Fall Conference
• /
• 2004.11a
• /
• pp.166-169
• /
• 2004

In this paper show how to design a redundant robot which is suitable for the multiple task without any constraints on the workspace. The implementation is possible by the rigid connection of a mobile robot and a task robot. Use a five joint articulated robot as the task robot; designed the 3 joint mobile robot for this usage. For a task execution assigned to the redundant robot, not only the task robot but the mobile robot should work in the coordinated way. therefore, a kinematic connection of the two robots should be cleary represented in a frame. And, also the dynamic interaction between the two robots needs to be analyzed. Clarified these issues considering the control of the redundant robot. Finally, demonstrate away of utilization of the redundancy as the cooperation between the mobile robot and the task robot to execute a common task.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.103-105
• /
• 2017

Hadoop MapReduce is a processing framework in which users' job can be efficiently processed in parallel and distributed ways on the Hadoop cluster. MapReduce task schedulers are used to select target nodes and assigns user's tasks to them. Previous schedulers cannot fully utilize resources of Hadoop cluster because they does not consider dynamic characteristics of cluster based on nodes' availability. To increase utilization of Hadoop cluster, this paper proposes a novel task assignment policy for MapReduce that assigns a job tasks to dynamic cluster efficiently by considering availability of each node.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.5
• /
• pp.64-72
• /
• 2004

An extension of interaction matrix formulation to the problem of system and disturbance identification for a plant that is corrupted by both process and output disturbances is presented. The teaming control develops controllers that learn to improve their performance at executing a given task, based on experience performing this task. The simplest forms of loaming control are based on the same concept as integral control, but operating in the domain of the repetitions of the task. This paper studies the use of such controllers in a decentralized system, such as a robot moving on the vertical plane with the controller for each link acting independently. The basic result of the paper is to show that stability and iterative precision of the learning controllers for all subsystems when the coupling between subsystems is turned off, assures stability of the decentralized teaming in the coupled system, provided that the sample time in the digital teaming controller is sufficiently short. The methods of teaming system are shown up for the iterative precision of each link.

• The Journal of Korean Physical Therapy
• /
• v.15 no.4
• /
• pp.65-81
• /
• 2003

The purpose of this study was to investigate the balance of stroke patient appling Task-Oriented Functional training program. Sixteen subjects were recruited from KeiMyung university Dongsan Medical Center inpatient satisfying requirement for this study. They were divided into Task-Oriented Functional training group and Conventional therapy group. They were measured by Sensory organization test and Motor control test items using EquiTest Version 8.0 machine before and after the test. The result of this study were to follow: 1. In Sensory organization test, there were no significant differences in static eguibrium score in both groups(p<.05). 2. In Sensory organization test, there were significant differences in dynamic eguibrium score in both groups(p<.05). 3. In Motor control test, there were significant differences in both groups(p<.05). 4. In exercise group, there were no significant differences in static Sensory organization test(p<.05). 5. In exercise group, there were significant differences in dynamic Sensory organization test(p<.05). 6. In exercise group, there were significant differences in Motor control test(p<.05). 7. In control group, there were no significant differences in static Sensory organization test(p<.05). 8. In control group, there were no significant differences in dynamic Sensory organization test(p<.05). 9. In control group, there were no significant differences in Motor control test(p<.05).