• Journal of Computing Science and Engineering
• /
• 제6권4호
• /
• pp.287-293
• /
• 2012

We introduce a methodology for the study of the application-level performance of time-sharing parallel jobs on a set of compute nodes in high performance clusters and report our findings. We assume that parallel jobs arriving at a cluster need to share a set of nodes with the jobs of other users, in that they must compete for processor time in a time-sharing manner and other limited resources such as memory and I/O in a space-sharing manner. Under the assumption, we developed a methodology to simulate job arrivals to a set of compute nodes, and gather and process performance data to calculate the percentage slowdown of parallel jobs. Our goal through this study is to identify a better combination of jobs that minimize performance degradations due to resource sharing and contention. Through our experiments, we found a couple of interesting behaviors for overlapped parallel jobs, which may be used to suggest alternative job allocation schemes aiming to reduce slowdowns that will inevitably result due to resource sharing on a high performance computing cluster. We suggest three job allocation strategies based on our empirical results and propose further studies of the results using a supercomputing facility at the San Diego Supercomputing Center.

• 복식문화연구
• /
• 제17권1호
• /
• pp.177-188
• /
• 2009

SCM as the important marketing strategy enhance the firm's efficiency and compatibility in global market environment such as global outsourcing. Firms adopted SCM realized the need to evaluate precisely the performance of SCM. In spite of importance of SCM, there was not much intention and research to measure SCM performance in textile fashion industry. Therefore, the purpose of this case study was to measure performance of supply chain management in textile fashion business using BSC(Balanced Score Card) to measure not only financial perspective but also non-financial perspectives such as customer perspective, internal business perspectives, financial perspective, and innovation & learning perspective. The questionnaire developed by the reviews of the literature was adopted for this study. The results of this study showed that SCM performance was enhanced from the point of customer perspective(cost, quality, time, service), financial perspective(cash cycle time, inventory turn over, inventory obsolescence, return on asset, return on investment, capacity utilization), and innovation & learning perspective(cost for human resource management, service for human resources). But there was same performance level regarding internal business perspective(lead time, cost for manufacturing process, product quality control, productive flexibility for time, quantity, and variety). Therefore, we should keep close relationship and two way communication among supply chain members to promote better SCM performance.

• Journal of Communications and Networks
• /
• 제5권2호
• /
• pp.135-140
• /
• 2003

In this paper, we investigate the error performance of a serially concatenated system using a nonrecursive convolutional code as the outer code and a recursive QPSK space-time trellis code as the inner code on quasi-static and rapid Rayleigh fading channels. At the receiver, we consider iterative decoding based on the maximum a posteriori (MAP) algorithm. The performance is evaluated by means of computer simulations and it is shown that better error performance can be obtained by using low complexity outer and/or inner codes and the Euclidean distance criterion based recursive space-time inner codes. We also obtain new systems with large number of trasmit and/or receive antennas providing good error performance.

• 한국수학교육학회지시리즈D:수학교육연구
• /
• 제16권3호
• /
• pp.155-176
• /
• 2012

It is an important issue to explore classroom environments which are conducive to developing students' mathematical performance. This study explores the effects of different classroom environments (solution-demand and corresponding-time setting) on mathematical performances. Fourteen and eighteen prospective teachers were required to prove a task under different conditions respectively: a) Cognitive demand of multiple-solution corresponding time of three hours, and b) Cognitive demand of a right solution corresponding time of 20 minutes. We used SOLO as the assessment tool for mathematical performance from quality perspective. Significant differences were found in the quantity and quality of mathematical performance. The regular environment focusing on speed and accuracy were found to be directly linked to low levels of performance. The findings above provide implications to the cognitive benefits of multiple-solution demand and corresponding time setting.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제10권1호
• /
• pp.1-9
• /
• 2010

In general, the performance of a component degrades as time goes by and failure of a component occurs when the performance degradation reaches a pre-specified level. It is difficult to obtain the failure time distribution data or the necessary number of failure data especially for the metal or machine part. Thus, a design of reliability qualification test based on performance distribution is more effective than failure time distribution. In this study, a performance-based reliability qualification test is developed and a numerical example is provided to illustrate the use of the developed reliability qualification test. This approach could be applied to many kinds of metal or machine part whose magnitude of strength could not be evaluated during at any random points but judgement can be made by only failure of the part. Besides, it is also possible that any parts which have a similar failure characteristics could be applicable to the developed reliability qualification test.

• 한국콘텐츠학회논문지
• /
• 제18권3호
• /
• pp.101-113
• /
• 2018

실시간으로 데이터를 획득하고 평가하는 것이 중요한 군용 점검 장비와 같은 시스템에서는 운영체제 레벨에서 실시간 처리가 가능해야 한다. 기술의 발전으로 기존 장비를 휴대용 단말로 대체하려는 요구가 있으나, 안드로이드가 적용된 휴대용 단말은 실시간성이 요구되는 시스템에 적합하지 않다. 안드로이드에서는 가비지 컬렉션을 통해 가용 메모리를 확보하는데, 이 작업이 수행되는 동안 다른 태스크가 중단되어 특정 태스크의 주기성을 보장할 수 없다. 본 논문에서는 이를 해결하기 위해 안드로이드의 가비지 컬렉션 실행을 제어하는 구조를 설계 및 구현하였다. 실시간 작업이 필요한 시간 동안 가비지 컬렉션을 제어하여 실시간 성능을 보장하며, 안드로이드에 실시간 성능 보장을 위한 RTiK을 적용하였다. 성능 평가를 위해 5ms 주기 태스크의 호출 주기를 측정하였으며, 제어 이전에는 태스크의 34.31%만 주기가 보장되었으나, 제어를 통해 98.18%의 태스크 주기가 만족되어 안드로이드에 실시간성을 제공하였다.

• 한국콘텐츠학회논문지
• /
• 제9권3호
• /
• pp.49-57
• /
• 2009

본 논문에서는 기존에 제시된 QoS 라우팅의 링크 상태 갱신 알고리즘의 성능을 향상시킬 수 있는 플로우 유지 시간을 기반으로 한 LSU 알고리즘을 제안하였다. 기존에 제시된 LSU 알고리즘은 네트워크 트래픽의 통계 정보를 기반으로 하여 LSU 메시지 전송 여부를 결정하는 반면, 제안하는 알고리즘은 플로우 유지 시간을 기반으로 하여 LSU 메시지의 개수를 감소시킬 수 있기 때문에 기존에 제안된 알고리즘에 모두 적용하여 사용할 수 있다. 제안하는 알고리즘은 짧은 시간 안에 반복적으로 발생되는 LSU 메시지의 개수를 감소시키기 위해 플로우의 유지 시간에 대한 통계적 정보를 사용하였다. 알고리즘의 성능 평가를 위해 기존에 제시된 다양한 LSU 알고리즘을 구현하여 본 논문에서 제안하는 알고리즘을 적용하였고 실제 네트워크와 유사한 MCI 네트워크상에서 라우팅 Blocking 확률과 링크 당 평균 LSU 메시지의 개수를 성능 평가 항목으로 하여 시뮬레이션을 수행하여 제시하는 알고리즘의 우수성을 확인하였다.

• Journal of Information Processing Systems
• /
• 제13권4호
• /
• pp.941-950
• /
• 2017

The round robin algorithm is regarded as one of the most efficient and effective CPU scheduling techniques in computing. It centres on the processing time required for a CPU to execute available jobs. Although there are other CPU scheduling algorithms based on processing time which use different criteria, the round robin algorithm has gained much popularity due to its optimal time-shared environment. The effectiveness of this algorithm depends strongly on the choice of time quantum. This paper presents a new effective round robin CPU scheduling algorithm. The effectiveness here lies in the fact that the proposed algorithm depends on a dynamically allocated time quantum in each round. Its performance is compared with both traditional and enhanced round robin algorithms, and the findings demonstrate an improved performance in terms of average waiting time, average turnaround time and context switching.

• International Journal of Contents
• /
• 제13권4호
• /
• pp.39-46
• /
• 2017

Due to resource constraints, most of Android application developers need to address potential performance problems during application development and maintenance. The coding styles and patterns of Android programming could often affect the execution time and energy efficiency which are utilized by the Android applications. Thus, it is necessary for application developers to apply performance-enhancing programming practices for mobile application development. This paper introduces performance-enhancing best practices for Android programming, and further, it evaluates the impact of these practices on the CPU time of the application. The original version with the performance-worsening code has been refactored to become an efficient version without changing its functionality. To demonstrate the efficiency of the proposed approach, each coding pattern was evaluated by measuring the CPU time under the controlled runtime environment. Furthermore, the Android applications were evaluated and compared via the CPU time of the original version, with that of the refactored version. These experimental results indicate that, by -using the proposed programming practices, the Android developer can develop performance-efficient mobile applications.

• Physical Therapy Rehabilitation Science
• /
• 제6권4호
• /
• pp.189-195
• /
• 2017

Objective: The purpose of this study was to examine the effects of real-time visual feedback weight shift training during golf swinging on golf performance. Design: Repeated-measures crossover design. Methods: Twenty-sixth amateur golfers were enrolled and randomly divided into two groups: The golf swing training with real-time feedback on weight shift (experimental group) swing training on the Wii balance board (WBB) by viewing the center of pressure (COP) trajectory on the WBB. All participants were assigned to the experimental group and the control group. The general golf swing training group (control group) performed on the ground. The golf performance was measured using a high-speed 3-dimensional camera sensor which analyses the shot distance, ball velocity, vertical launch angle, horizontal launch angle, back spin velocity and side spin velocity. The COP trajectory was assessed during 10 practice sessions and the mean was used. The golf performance measurement was repeated three times and its mean value was used. The assessment and training were performed at 24-hour intervals. Results: After training sessions, the change in shot distance, ball velocity, and horizontal launch angle pre- and post-training were significantly different when using the driver and iron clubs in the experimental group (p<0.05). The interaction time${\times}$group and time${\times}$club were not significant for all variables. Conclusions: In this study, real-time feedback training using real-time feedback on weight shifting improves golf shot distance and accuracy, which will be effective in increasing golf performance. In addition, it can be used as an index for golf player ability.