• Journal of the Korea Society of Computer and Information
• /
• v.21 no.1
• /
• pp.125-129
• /
• 2016

In this paper, we propose an efficient dynamic workload balancing strategy which improves the performance of high-performance computing system. The key idea of this dynamic workload balancing strategy is to minimize execution time of each job and to maximize the system throughput by effectively using system resource such as CPU, memory. Also, this strategy dynamically allocates job by considering demanded memory size of executing job and workload status of each node. If an overload node occurs due to allocated job, the proposed scheme migrates job, executing in overload nodes, to another free nodes and reduces the waiting time and execution time of job by balancing workload of each node. Through simulation, we show that the proposed dynamic workload balancing strategy based on CPU, memory improves the performance of high-performance computing system compared to previous strategies.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2003.11a
• /
• pp.27-34
• /
• 2003

A distributed server cluster system is a cost-effective system to provide a service application for clients with reliable, scalable, available, and fault-tolerant features. In order to provide high quality services, it is necessary to evaluate service performances, tune the server system, and analyze performances. In this paper, we propose a simulator to generate workloads based on statistic configuration according to estimated application traffics, apply workload scheduling algorithms, and evaluate the simulation results. We introduce the simulator design modelling and architecture. By using flexible parameters, the simulator is able to generate various patterns of workloads with different statistics, and configure system environments such as the number of server nodes, system resources considered, and their capacities. With this simulator, we introduce two scenarios: one is to find appropriate thresholds for the best performance of cluster system, and the other is to find the suitable scheduling algorithm for workload characteristics of service applications.

• Journal of Internet Computing and Services
• /
• v.15 no.3
• /
• pp.109-116
• /
• 2014

This study proposed new instance selection using neural network with weighted fuzzy membership functions(NEWFM) based on Takagi-Sugeno(T-S) fuzzy model to improve the classification performance. The proposed instance selection adopted weighted average defuzzification of the T-S fuzzy model and an interval selection, same as the confidence interval in a normal distribution used in statistics. In order to evaluate the classification performance of the proposed instance selection, the results were compared with depending on whether to use instance selection from the case study. The classification performances of depending on whether to use instance selection show 77.33% and 78.19%, respectively. Also, to show the difference between the classification performance of depending on whether to use instance selection, a statistics methodology, McNemar test, was used. The test results showed that the instance selection was superior to no instance selection as the significance level was lower than 0.05.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1571-1576
• /
• 2005

DCS has many processing components and various communication elements. And its communication delay characteristic is affected diverse operating situation and context. Especially, binding signal which traversed from one control-node to another control-node undergo all sort of delay conditions. So its delay value has large deviation with the lapse of time, and the measurement of delay statistics during long time is very difficult by using general oscilloscope or other normal instruments. This thesis introduces the design and implementation of PC-based BDAS(Binding Delay Analysis System) System developed to overcomes these hardships. The system has signal-generator, IO-card, data-acquisition module, delay-calculation and analyzer module, those are implemented on industrial standard PC/Ethernet hardware and Windows/Linux platforms. This system can detect accurate whole-system-wide delay time including io, control processing and network delay, in the resolution of msec unit, and can analyze each channel's delay-historic data which is maintained by realtime database. So, this system has strong points of open system architecture, for example, user-friendly environment, low cost, high compatibility, simplicity of maintenance and high extension ability. Of all things, the measuring capability of long-time delay-statistics obtained through historic-DB make the system more valuable and useful, which function is essential to analyze accurate delay performance of DCS system. Using this system, the verification of delay performance of DCS for nuclear power plants is succeeded in KNICS(Korea Nuclear Instrumentation & Control System) projects

• The Korean Journal of Applied Statistics
• /
• v.26 no.3
• /
• pp.441-452
• /
• 2013

This study analyzed the relationship between key film and a box office record success factors based on movies released in the first quarter of 2013 in Korea. An over-fitting problem can happen if there are too many explanatory variables inserted to regression model; in addition, there is a risk that the estimator is instable when there is multi-collinearity among the explanatory variables. For this reason, optimal variable selection based on high explanatory variables in box-office performance is of importance. Among the numerous ways to select variables, LASSO estimation applied by a generalized linear model has the smallest prediction error that can efficiently and quickly find variables with the highest explanatory power to box-office performance in order.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.1
• /
• pp.113-120
• /
• 2012

When we evaluate prospective students in the interview process, we have to implement a system in which each student can be fairly judged. This process, the standardization of the scores which the interviewers have produced based on a student's performance, is implemented to ensure that each student receives a score that objectively translates one's performance. Although we don't know exactly how effective the standardization is in many different cases, we have researched which standardization methods are most stable and have minimum risks among the four methods such as STD, Range, MAD and IQR. These methods use scales such as standard deviation, range, maximum median and interquartile range.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.1
• /
• pp.143-150
• /
• 2012

Based on the student evaluation of teaching from 'A' university where students are graded by relative performance, we conducted a research on the effect of course characteristics in student evaluation of teaching. The results show that the score of student evaluation of teaching seems to be higher for those classes with more proportion of male students, higher grades, and smaller class sizes. From these results, we suggest an evaluation model which can control the effect of grade and sex. Also we illustrate the performance of the evaluation model by using a case study.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.11 no.4
• /
• pp.227-236
• /
• 1986

In this paper, an integrated voice/data packet network with window flow control is modeled by a colsed multichain queueing system, and its performance is analyzed by the mean value analysis method. Particularly, for the analysis of a packet network having various kinds of messages with different priority classes, we introduce an approach based on the mean value analysis and the concept of effective capacity. By the mathematical analysis and computer simulation, we obtain the following network statistics in the steady state: Mean buffer occupancy at each node, utilization of link throughput of a virtual channel, and mean delay time of each message. Our iterative analysis method can predict the link data status in most cases within about 10 percent of accurady, and the statistics of voice messages and external data within 5 percent as compared to simulation results.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.19 no.1
• /
• pp.113-130
• /
• 2008

Performance indicators for academic libraries in Korea suggested in this research cover three areas, that is, collection, budget, and human resources, which consists of three subdivisions, respectively. Performance indicators and action plans in this research were prepared, based on the official documents which universities in Korea formally submitted to the Ministry of Education and Human Resources, and the data published by the related institutions. Performance indicators and action plans are intended to draw attentions of university administrators and headquarters, but do not require librarians to prepare for evaluation. The better results would be expected with the development and application of both quantitative and qualitative indicators for information services by standardizing and unifying the process of collecting statistics. It is also expected that periodic and continuing evaluation of academic libraries could help government and universities have a better understanding of and appreciate the significance of academic libraries.

• Journal of the Korean Statistical Society
• /
• v.30 no.3
• /
• pp.453-465
• /
• 2001

It is often the case that one wants to estimate parameters of the distribution which follows certain parametric model, while the dta are contaminated. it is well known that the maximum likelihood estimators are not robust to contamination. Basuet al.(1998) proposed a robust method called the minimum density power divergence estimation. In this paper, we investigate data-driven selection of the tuning parameter $\alpha$ in the minimum density power divergence estimation. A criterion is proposed and its performance is studied through the simulation. The simulation includes three cases of estimation problem.