• 한국시뮬레이션학회논문지
• /
• 제6권1호
• /
• pp.1-13
• /
• 1997

In this paper, we present a partitioning algorithm for distributed simulation of DEVS (Discrete Event System Specification) models. To preserve concurrency inherent in models, the proposed algorithm utilizes the structural information of models. Through benchmark simulation experiments, we show that the proposed algorithm can generate good partitions.

• 한국정보컨버전스학회논문지
• /
• 제7권2호
• /
• pp.49-57
• /
• 2014

최근 대량의 SNS(Social Network Service) 데이터로부터 유용한 정보를 추출하고 사용자의 진의 정보를 평가하기 위한 오피니언 마이닝(opinion mning)이 소개되고 있다. 오피니언 마이닝은 대량의 SNS 데이터로부터 빠른 기간 내에 데이터를 수집하고 분석하여 목적에 적합한 정보를 추출하는 효율적인 기법이 필요하다. SNS에서 발생되는 다양한 비정형 데이터로부터 감성정보를 추출하기 위해, 본 논문에서는 하둡(Hadoop) 시스템 기반의 병렬적 HDFS(Hadoop Distributed File System)와 맵리듀스(MapReduce) 기반 감성분석 함수를 제안한다. 실험결과로 제안한 시스템과 함수는 데이터 수집과 적재시간에 대해 O(n)보다 빠르게 처리하며, 메모리와 CPU 자원에 대해 안정적인 부하분산이 이루어지는 것을 확인하였다.

• 한국지능시스템학회논문지
• /
• 제15권3호
• /
• pp.324-329
• /
• 2005

본 논문에서는 시간 지연을 갖는 이산 시간 비선형 시스템을 $H\infty$ 의미에서 안정하게 하는 정적 출력 제한 퍼지 제어기 설계를 제시한다. 먼저 대상이 되는 비선형 시스템은 Takagi-Sugeno 퍼지 모델로 표현 되어진다. 그리고 parallel distributed compensation technique을 이용하여 퍼지 제어기의 형태를 만든다. 하나의 Lyapunov 함수를 정하여서 폐루프 시스템의 전역 점근적 안정성과 외란에 대한 강인성을 bilinear matrix inequality 형태로 제시한다. 그리고 합동변환법과 동질성 변환법을 통해 이것을 선형 행렬 부등식 (linear matrix inequality) 으로 표현한다. 제안된 방법의 효율성과 가능성을 보여주기 위해 한 예제를 포함한다.

• 한국콘텐츠학회논문지
• /
• 제8권12호
• /
• pp.441-448
• /
• 2008

분자 도킹 실험은 일반적으로 계산 량이 매우 많아 슈퍼 컴퓨팅 파워를 요구하는 실험이다. 따라서 시간이 많이 소요되기 때문에 일반적으로 CPU가 탑재된 컴퓨터를 여러 대 묶어서 사용하는 분산 환경 혹은 그리드 환경에서 실험을 수행하고 있다. 이와 같은 실험 환경은 시간적, 공간적 제약성이 많아 일반적으로 과학자들이 접근하기가 어렵다. 그래서 근래에는 많은 CPU를 사용하기 보다는 월등히 성능이 높은 GPU를 병렬 화하여 과학 분야에 계산하는 연구가 매우 활발히 이루어지고 있는 추세이다. CUDA는 병렬 GPU 프로그래밍을 가능하게 하는 공개 기술이다. 본 논문에서는 이러한 CUDA 기술을 사용하여 분자 도킹 실험을 할 수 있는 시스템을 제안한다. 또한, 분자 도킹 실험에 있어서 중요한 에너지 최소화 계산을 병렬 화하는 알고리즘을 제안한다. 이와 같은 실험을 검증하기 위해 본 논문에서는 일반적인 CPU에서 분자 도킹 실험 시간과 본 논문에서 제안한 병렬 CPU 기반의 분자 도킹 시간을 비교 분석 하였다.

• 산업기술연구
• /
• 제10권
• /
• pp.15-20
• /
• 1990

A parallel processing algorithm for discrete Kalman filter, which is one of the most commonly used filtering technique in modern control, signal processing, and communication, is proposed. Previously proposed parallel algorithms to decrease the number of computations needed in the Kalman filter are the hierachical structures by distributed processing of measurements, or the systolic structures to disperse the computational burden. In this paper, a new parallel Kalman filter employing a structure similar to recursive doubling is proposed. Estimated values of state variables by the new algorithm converge with two times faster data processing speed than that of the conventional Kalman filter. Moreover it maintains the optimality of the conventional Kalman filter.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2005년도 추계 학술대회논문집
• /
• pp.157-162
• /
• 2005

Parallel implementation and performance assessment of the grid assembly in a structured chimera grid approach is studied. The grid assembly process, involving hole cutting and searching donor, is parallelized on the PC cluster. A message passing programming model based on the MPI library is implemented using the single program multiple data(SPMD) paradigm. The coarse-grained communication is optimized with the minimized memory allocation because that the parallel grid assembly can access the decomposed geometry data in other processors by only message passing in the distributed memory system such as a PC cluster. The grid assembly workload is based on the static load balancing tied to flow solver. A goal of this work is a development of parallelized grid assembly that is suited for handling multiple moving body problems with large grid size.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1990년도 하계학술대회 논문집
• /
• pp.64-67
• /
• 1990

A parallel processing algorithm for discrete Kalman filter, which is one of the most commonly used filtering technique in modern control, signal processing, and communication. is proposed. Previously proposed parallel algorithms to decrease the number of computations needed in the Kalman filter are the hierachical structures by distributed processing of measurements, or the systolic structures to disperse the computational burden. In this paper, a new parallel Kalman filter employing a structure similar to recursive doubling is proposed. Estimated values of state variables by the new algorithm converge with two times faster data processing speed than that of the conventional Kalman filter. Moreover it maintains the optimality of the conventional Kalman filter.

• Structural Engineering and Mechanics
• /
• 제32권6호
• /
• pp.787-809
• /
• 2009

Recent interest in the use of wireless sensor networks for structural health monitoring (SHM) is mainly due to their low implementation costs and potential to measure the responses of a structure at unprecedented spatial resolution. Approaches capable of detecting damage using distributed processing must be developed in parallel with this technology to significantly reduce the power consumption and communication bandwidth requirements of the sensor platforms. In this investigation, a damage detection system based on a distributed processing approach is proposed and experimentally validated using a wireless sensor network deployed on two laboratory structures. In this distributed approach, on-board processing capabilities of the wireless sensor are exploited to significantly reduce the communication load and power consumption. The Damage Location Assurance Criterion (DLAC) is used for localizing damage. Processing of the raw data is conducted at the sensor level, and a reduced data set is transmitted to the base station for decision-making. The results indicate that this distributed implementation can be used to successfully detect and localize regions of damage in a structure. To further support the experimental results obtained, the capabilities of the proposed system were tested through a series of numerical simulations with an expanded set of damage scenarios.

• International Journal of Reliability and Applications
• /
• 제14권1호
• /
• pp.11-26
• /
• 2013

The availability equivalence factors of a general repairable series-parallel system is discussed in this paper based on the availability function of the system. The system components are assumed to be repairable and independent but not identical. The life and repair times of the system components are exponentially distributed with different parameters. Two types of availability equivalent factors of the system are derived. The results derived in this paper generalizes those given in the literature. A case study is introduced to illustrate how the idea of this work can be applied.

• International Journal of Reliability and Applications
• /
• 제16권1호
• /
• pp.1-13
• /
• 2015

In this paper, we investigate reliability and availability of repairable systems with two types of failure. The first one is to one unit and the second one is to M units in parallel structure. Let failure rate and repair rate of [type1, type2] components are assumed to be exponentially distributed. The expressions of availability and reliability characteristics such as the system reliability and the mean time to failure are derived for two systems. We used several cases to analyze graphically the effect of various system parameters on the reliability system and availability system.