• 제목/요약/키워드: parallel computers

검색결과 141건 처리시간 0.022초

분산컴퓨팅 환경에서 공력 설계최적화의 효율성 연구 (A STUDY ON THE EFFICIENCY OF AERODYNAMIC DESIGN OPTIMIZATION USING DISTRIBUTED COMPUTATION)

  • 김양준;정현주;김태승;조창열
    • 한국전산유체공학회:학술대회논문집
    • /
    • 한국전산유체공학회 2005년도 추계 학술대회논문집
    • /
    • pp.163-167
    • /
    • 2005
  • A research to evaluate efficiency of design optimization was performed for aerodynamic design optimization problem in distributed computing environment. The aerodynamic analyses which take most of computational work during design optimization were divided into several jobs and allocated to associated PC clients through network. This is not a parallel process based on domain decomposition rather than a simultaneous distributed-analyses process using network-distributed computers. GBOM(gradient-based optimization method), SAO(Sequential Approximate Optimization) and RSM(Response Surface Method) were implemented to perform design optimization of transonic airfoil and to evaluate their efficiencies. One dimensional minimization followed by direction search involved in the GBOM was found an obstacle against improving efficiency of the design process in distributed computing environment. The SAO was found quite suitable for the distributed computing environment even it has a handicap of local search. The RSM is apparently the fittest for distributed computing environment, but additional trial and error works needed to enhance the reliability of the approximation model are annoying and time-consuming so that they often impair the automatic capability of design optimization and also deteriorate efficiency from the practical point of view.

  • PDF

프로세스 결함 검출을 위한 OGSA 기반 그리드 서비스의 설계 및 구현 (A Grid Service based on OGSA for Process Fault Detection)

  • 강윤희
    • 한국콘텐츠학회:학술대회논문집
    • /
    • 한국콘텐츠학회 2004년도 추계 종합학술대회 논문집
    • /
    • pp.314-317
    • /
    • 2004
  • 네트워크 및 소프트웨어 하부구조의 진보에 의해 이형의 컴퓨팅 자원 집합으로 구성된 환경에서 그리드 컴퓨팅 기술은 확산되고 있다. 그리드 컴퓨팅은 광대역으로 연결되어진 분산 컴퓨터들의 조정에 의한 사용을 요구한다. 그리드 컴퓨팅을 위한 다양한 시스템과 그리드 응용 서비스의 증가에 따라 그리드 운영환경은 네트워크 분리 및 노드 결함 등에 의해 높은 결함율을 가질 수 있다. 결함 검출은 그리드 응용 시스템의 강건성을 위해 시스템 설계 및 구현시의 필수적인 요소이다. 본 논문에서는 낮은 네트워크 트래픽 환경에서 높은 신뢰성 정공을 위한 OGSA(Open Grid Service Architecture) 기반의 자원 결함 검출 서비스를 제안한다.

  • PDF

UHF 대역 무선 프린터 공유 인터페이스의 설계 (The Design of Wireless Printer Interfaces for UHF band)

  • 강영석;김기래
    • 한국정보통신학회:학술대회논문집
    • /
    • 한국해양정보통신학회 2000년도 춘계종합학술대회
    • /
    • pp.280-283
    • /
    • 2000
  • 여러 대의 컴퓨터를 사용하는 경우 프린터의 공유 방법에 관한 문제가 제기된다. LAN에 의한 프린터공유는 보편적이나, LAN이 설치되지 않은 환경에서는 프린터 공유기를 이용하거나 프린터를 이동시켜 인쇄를 하여야 한다. 본 연구에서는 컴퓨터와 프린터 사이에 무선으로 데이터를 송수신 한 수 있는 무선 인터페이스 프린터 공유기를 개발하였다. 이 장치를 이용함으로써 프린터의 이동 없이 최대 50미터 반경내의 다수의 컴퓨터와 시분할 방식의 공유가 가능하다. 본 장치는 프린터와 컴퓨터에 UHF 송수신 인터페이스 모듈을 각각 설치하여 데이터를 전송하며, 최대 전송속도는 9600bps이다.

  • PDF

분산컴퓨팅 환경에서 공력 설계최적화의 효율성 연구 (A STUDY ON THE EFFICIENCY OF AERODYNAMIC DESIGN OPTIMIZATION IN DISTRIBUTED COMPUTING ENVIRONMENT)

  • 김양준;정현주;김태승;손창호;조창열
    • 한국전산유체공학회지
    • /
    • 제11권2호
    • /
    • pp.19-24
    • /
    • 2006
  • A research to evaluate the efficiency of design optimization was carried out for aerodynamic design optimization problem in distributed computing environment. The aerodynamic analyses which take most of computational work during design optimization were divided into several jobs and allocated to associated PC clients through network. This is not a parallel process based on domain decomposition in a single analysis rather than a simultaneous distributed-analyses using network-distributed computers. GBOM(gradient-based optimization method), SAO(Sequential Approximate Optimization) and RSM(Response Surface Method) were implemented to perform design optimization of transonic airfoils and evaluate their efficiencies. dimensional minimization followed by direction search involved in the GBOM was found an obstacle against improving efficiency of the design process in the present distributed computing system. The SAO was found fairly suitable for the distributed computing environment even it has a handicap of local search. The RSM is apparently the most efficient algorithm in the present distributed computing environment, but additional trial and error works needed to enhance the reliability of the approximation model deteriorate its efficiency from the practical point of view.

휴대용 컴퓨터 내에 실장된 강제공랭 모듈 주위의 유체유동과 온도분포 (Fluid Flow and Temperature Distribution Around a Surface-Mounted Module Cooled by Forced Air Flow in a Portable Personal Computers)

  • 박상희;신대종
    • 대한기계학회논문집B
    • /
    • 제28권2호
    • /
    • pp.238-246
    • /
    • 2004
  • This paper reports an experimental study around a module about forced air flow by blower (35${\times}$35${\times}$6㎣) in a portable personal computer model(200${\times}$235${\times}$10㎣). Experimental report is to know three data to investigate thermal resistance, adiabatic wall temperature and visualized fluid flow around the module by combination of the moving number and the arrangement method of blower. The channel inlet flow velocity has been varied between 0.26, 0.52 and 0.78㎧, and input power ( $Q_{p}$) to the module is 4W. To investigate thermal resistance. the heated module is mounted on two boards(110${\times}$110${\times}$1.2㎣, k=20.73, 0.494W/ $m^{\circ}C$) in parallel-plate channel to forced air flow. The temperature distribution were visualized by heated module on acrylic board(k=0.262W/ $m^{\circ}C$) using liquid crystal film. Fluid flow around the module were visualized using particle image velocimetry system.

반도체 Bump 검사를 위한 백색광 주사 간섭계의 고속화 (A High-Speed White-Light Scanning Interferometer for Bump Inspection of Semiconductor Manufacture)

  • 고국원;심재환;김민영
    • 한국정밀공학회지
    • /
    • 제30권7호
    • /
    • pp.702-708
    • /
    • 2013
  • The white-light scanning interferometer (WSI) is an effective optical measurement system for high-precision industries (e.g., flat-panel display and electronics packaging manufacturers) and semiconductor manufacturing industries. Its major disadvantages include a slow image-capturing speed for interferogram acquisition and a high computational cost for peak-detection on the acquired interferogram. Here, a WSI system is proposed for the semiconductor inspection process. The new imaging acquisition technique uses an 'on-the-fly' imaging system. During the vertical scanning motion of the WSI, interference fringe images are sequentially acquired at a series of pre-defined lens positions, without conventional stepwise motions. To reduce the calculation time, a parallel computing method is used to link multiple personal computers (PCs). Experiments were performed to evaluate the proposed high-speed WSI system.

Analysis of reinforced concrete corbel beams using Strut and Tie models

  • Parol, Jafarali;Al-Qazweeni, Jamal;Salam, Safaa Abdul
    • Computers and Concrete
    • /
    • 제21권1호
    • /
    • pp.95-102
    • /
    • 2018
  • Reinforced concrete corbel beams (span to depth ratio of a corbel is less than one) are designed with primary reinforcement bars to account for bending moment and with the secondary reinforcement placed parallel to the primary reinforcement (shear stirrups) to resist shear force. It is interesting to note that most of the available analytical procedures employ empirical formulas for the analysis of reinforced concrete corbels. In the present work, a generalized and a simple strut and tie models were employed for the analysis of reinforced corbel beams. The models were benchmarked against experimental results available in the literature. It was shown here that increase of shear stirrups increases the load carrying capacity of reinforced concrete corbel beams. The effect of horizontal load on the load carrying capacity of the corbel beams has also been examined in the present paper. It is observed from the strut and tie models that the resistance of the corbel beam subjected to combined horizontal and vertical load did not change with increase in shear stirrups if the failure of the corbel is limited by concrete crushing. In other words, the load carrying capacity was independent of the horizontal load when failure of the beam occurred due to concrete crushing.

Prediction of creep in concrete using genetic programming hybridized with ANN

  • Hodhod, Osama A.;Said, Tamer E.;Ataya, Abdulaziz M.
    • Computers and Concrete
    • /
    • 제21권5호
    • /
    • pp.513-523
    • /
    • 2018
  • Time dependent strain due to creep is a significant factor in structural design. Multi-gene genetic programming (MGGP) and artificial neural network (ANN) are used to develop two models for prediction of creep compliance in concrete. The first model was developed by MGGP technique and the second model by hybridized MGGP-ANN. In the MGGP-ANN, the ANN is working in parallel with MGGP to predict errors in MGGP model. A total of 187 experimental data sets that contain 4242 data points are filtered from the NU-ITI database. These data are used in developing the MGGP and MGGP-ANN models. These models contain six input variables which are: average compressive strength at 28 days, relative humidity, volume to surface ratio, cement type, age at start of loading and age at the creep measurement. Practical equation based on MGGP was developed. A parametric study carried out with a group of hypothetical data generated among the range of data used to check the generalization ability of MGGP and MGGP-ANN models. To confirm validity of MGGP and MGGP-ANN models; two creep prediction code models (ACI209 and CEB), two empirical models (B3 and GL 2000) are used to compare their results with NU-ITI database.

교육 분야 응용을 위한 가상 현미경 시스템 (A Virtual Microscope System for Educational Applications)

  • 조승호
    • 정보처리학회논문지D
    • /
    • 제10D권1호
    • /
    • pp.117-124
    • /
    • 2003
  • 본 논문에 의해 구현된 시스템은 광학 현미경을 통해 캡춰된 조직 샘플들에 대한 데이터들을 분산/병렬 시스템 상에 분한 저장한다. 사용자들은 컴퓨터 상에서 마치 현미경을 사용하듯이 이들 이미지들을 관찰할 수 있다. 이 시스템은 고객 서버 모델에 의거 고객, 조정자 노드, 데이터 관리자 노드로서 구성되고, 메시지를 통해 데이터를 송수신한다. 본 연구에서는 이미지 표시나 텍스트 주석 등 교육용 응용에 필요한 기능들을 갖춘 이미지 검색용 고객 프로그램을 구현하였고, 세 요소간 통신 규약을 정의하였다. 또한 대용량 데이터들을 저장하는 테이프 장치 도입을 위한 실험을 수행하였으며, 이러한 실험은 데이터 분할 및 인덱싱 기법에 의해 성능 향상을 나타내었다.

인터넷 기반 근사 형상최적설계의 분산처리 (Distributed Process of Approximate Shape Optimization Based on the Internet)

  • 임오강;최은호;김우현
    • 한국전산구조공학회논문집
    • /
    • 제21권4호
    • /
    • pp.317-324
    • /
    • 2008
  • 대형 구조물에 대한 최적설계를 고려할 때 구조해석에 많은 시간과 노력이 소비된다. 한대의 개인용 컴퓨터에 의한 대형 구조물의 구조해석은 대용량의 기억장치와 많은 계산시간이 요구되므로 반복적 해석이 필요한 대형 구조물의 설계에 효율적으로 이용되기 어렵다. 따라서 본 논문에서는 이러한 문제의 대안으로 인터넷이 연결된 다수의 개인용 컴퓨터들로 고성능 병렬연산시스템을 구성하여 구조해석을 분산 처리하여 계산시간을 절감하였다. 아울러 반응표면의 근사를 위해 요구되는 구조해석을 상용 구조해석 어플리케이션으로 해결할 수 있다면 상용성이 확보되어 일반 구조물에 대하여도 반응표면법을 이용한 최적설계를 수행할 수 있을 것이다.