• 한국컴퓨터정보학회논문지
• /
• 제19권10호
• /
• pp.1-12
• /
• 2014

본 논문에서는 단일 명령어, 다중 데이터 처리 기반의 매니코어 프로세서를 이용하여 높은 계산량이 요구되는 차감 클러스터링 알고리즘을 병렬 구현하고 성능을 향상시킨다. 또한 차감 클러스터링 알고리즘을 위한 최적의 매니코어 프로서서 구조를 선택하기 위해 다섯 가지의 프로세싱 엘리먼트 (processing element, PE) 구조 (PEs=16, 64, 256, 1,024, 4,096)를 모델링하고, 각 PE구조에 대해 실행시간 및 에너지 효율을 측정한다. 두 가지 의료 영상 및 각 영상의 세 가지 해상도(($128{\times}128$, $256{\times}256$, $512{\times}512$)를 이용하여 모의 실험한 결과, 모든 경우에 대해 PEs=4,096구조에서 최고의 성능 및 에너지 효율을 보였다.

• Computers and Concrete
• /
• 제13권2호
• /
• pp.271-290
• /
• 2014

In this study, structural irregularities in plan, which has a considerable effect on earthquake behavior of buildings, have been investigated in detail based on Turkish Earthquake Code 2007. The study consists of six main parametric models and a total of 144 sub-models that are grouped based on RC structural systems such as frame, frame + rigid core, frame with shear wall, and frame with shear wall + rigid core. All models are designed to have both symmetrical plan geometry and regular rigidity distribution. Changes in the earthquake behavior of buildings were evaluated according to the number of storeys, number of axes and the configuration of structural elements. Many findings are obtained and assessed as a result of the analysis for each structural irregularity. The study shows that structural irregularities can be observed in completely symmetric buildings in terms of plan geometry and rigidity distribution.

• 한국조경학회지
• /
• 제35권5호
• /
• pp.56-63
• /
• 2007

This study used the nearest feature model to connect forest patches within the western Geum River Basin. Due to many different forest patch sizes, 3 alternative methods were tested to determine the best way to establish an ecological network with forest patches. Alternative 1 used all forest patches to determine whether patches were large enough. Alternative 2 used forest patches over 10 ha in size. Alternative 3 used natural conservation indices to select forest patches containing better qualities in the natural conservation level. As a result 635 out of 724 patches of over 10 ha were selected for comparison. Alternative 1 showed that forest patches of less than 10 ha were outliers interrupting the establishment of the ecological network. They generated an unnecessary ecological network to link core areas to comparison features. The ecological network was improved by using forest patches greater than 10 ha in size(Alternative 2). Each comparison feature was much more hierarchically connected to core areas in Alternative 2 than in Alternative 1. Forest patches filtered by natural conservation indices were useful for obtaining the best ecological network. Alternative 3 clearly showed the connections in the ecological network between core areas and forest.

• 정보통신설비학회논문지
• /
• 제10권1호
• /
• pp.7-13
• /
• 2011

Matrix multiplication is a fundamental operation of linear algebra and arises in many areas of science and engineering. This paper introduces an efficient parallel matrix multiplication scheme on N ${\times}$ N mesh-connected SIMD array processor, called multiple hierarchical SIMD architecture (HMSA). The architectural characteristic of HMSA is the hierarchically structured control units which consist of a global control unit, N local control units configured diagonally, and $N^2$ processing elements (PEs) arranged in an N ${\times}$ N array. PEs are communicating through local buses connecting four adjacent neighbor PEs in mesh-torus networks and global buses running across the rows and columns called horizontal buses and vertical buses, respectively. This architecture enables HMSA to have the features of diagonally indexed concurrent broadcast and the accessibility to either rows (row control mode) or columns (column control mode) of 2D array PEs alternately. An algorithmic mapping method is used for performance evaluation by mapping matrix multiplication on the proposed architecture. The asymptotic time complexities of them are evaluated and the result shows that paralle matrix multiplication on HMSA can provide significant performance improvement.

• 정보처리학회논문지B
• /
• 제18B권4호
• /
• pp.173-180
• /
• 2011

본 논문은 GPU(Graphics Processing Unit) 에서 CUDA(Compute Unified Device Architecture)를 사용하여 실시간으로 객체를 분할하는 방법을 소개한다. 최근에 감시 시스템, 오브젝트 추적, 모션 분석 등의 많은 응용 프로그램들은 실시간 처리가 요구된다. 이러한 단계의 선행부분인 객체 분할 기법은 기존 CPU 기반의 시스템으로는 실시간 처리에 제약이 발생한다. NVIDIA에서는 Parallel Processing for General Computation 을 위해 그래픽 하드웨어 제약을 개선한 CUDA platform을 제공하고 있다. 본 논문에서는 객체 추출 단계에 대표적인 적응적 가우시안 혼합 배경 모델링(Adaptive Gaussian Mixture Background Modeling) 알고리즘과 Classification 기법으로 사용되는 CCL (Connected Component Labeling) 알고리즘을 적용하였다. 본 논문은 2.4GHz를 갖는 Core2 Quad 프로세서와 비교하여 평가하였고 그 결과 3~4배 이상의 성능향상을 확인할 수 있었다.

• 대한건축학회논문집:계획계
• /
• 제34권7호
• /
• pp.89-98
• /
• 2018

Koolhaas' perspective on urbanism differs from the modern urbanism and typological urbanism. The Melun-Senart masterplan, La $D{\acute{e}}fense$ masterplan shows unique characteristic different from that of conventional urbanism. The roots of this creative approach can be found in the traits of his research with O.M. Ungers back in the 1970s. Koolhaas and Ungers have collaborated intimately from 1972 to 1977 to work on urban projects. This collaboration reaches its cilmax in their project City in the City, where many of Koolhaas' core concepts such as archipelago, void/solid, plurality, infrastructure, congestion, and social condenser are introduced. This thesis will explore the development of these concepts in their collaboration and shed new light on how this period has made a transition into Koolhaas' perspective on urbanism and architecture as well as his works. The purpose of this study is to investigate these core concepts of Koolhaas in City in the City, and to find the development and the meaning of these concepts in his projects.

• 한국게임학회 논문지
• /
• 제2권1호
• /
• pp.30-35
• /
• 2002

The deployment of multicast communication services in the Internet is expected to lead a stable packet transfer even in heavy traffic as in On-Line Game environment. The Core Based Tree scheme among many multicast protocols is the most popular and suggested recently. However, CBT exhibit two major deficiencies such as traffic concentration or poor core placement problem. So, measuring the bottleneck link bandwidth along a path is important for understanding the performance of multicast. We propose not only a definition of CBT's core link state that Steady-State(SS), Normal-State(NS) and Bottleneck State(BS) according to the estimation link speed rate, but also the changeover of multicast routing scheme for traffic overload. In addition, we introduce anycast routing tree, a efficient architecture for construct shard multicast trees.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2005년도 제36회 하계학술대회 논문집 D
• /
• pp.2936-2938
• /
• 2005

The deployment of multicast communication services in the internet is expected to lead a stable packet transfer even in heavy traffic as in network system environment. The core based tree scheme among many multicast protocols is the most popular and suggested recently. However, CBT exhibit two major deficiencies such as traffic concentration or poor core placement problem. so, measuring the bottleneck link bandwidth along a path is important for understanding th performance of multicast. We propose not only a definition of CBT's core link state that Steady-State, Normal-State and Bottleneck State according to the estimation link speed rate, but also the changeover of multicast routing scheme for traffic overload. In addition, we introduce anycast routing tree, a efficient architecture for construst shard multicast trees.

• 게임&엔터테인먼트 논문지
• /
• 제2권1호
• /
• pp.8-16
• /
• 2006

인터넷에서 게임 트래픽과 같은 대용량 데이터의 부하를 분산하고 안정적인 전송을 위해 멀티캐스트 라우팅 프로토콜을 사용한다. 다양한 멀티캐스트 알고리즘 중 최근 CBT(Core Based Tree) 방식이 널리 사용되고 있으나 코어 라우터로 트래픽 집중되는 병목현상과 코어라우터의 위치에 따라 푸어 코어(Poor core)현상이 발생한다. 본 논문에서는 코어 라우터의 병목현상을 초래하는 링크 레이트(Link Rate)에 따라 코어 라우터의 상태를 SS(Steady State), NS(Normal State), BS(Bottleneck State)로 구분하였으며, 코어 라우터의 과부하에 따라 멀티캐스트 라우팅 전략을 CBT에서 Anycast로 전환하는 방식을 제안하였다. 온라인 게임에서 교환되는 주요 패킷의 크기에 준하여 두 가지 라우팅 방식을 비교하고 트래픽의 증가에 따라 Anycast 라우팅 방식의 성능개선을 보였다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2014년도 춘계학술발표대회
• /
• pp.87-89
• /
• 2014

Aho-Corasick (AC) 알고리즘은 실시간 성능을 요하는 많은 응용 분야에 적용되는 스트링 매칭 알고리즘으로서, 한번에 여러 개의 패턴들을 동시에 매칭시키는 것이 가능하다. 본 논문에서는 Intel 의 Many Integrated Core (MICO 아키텍쳐인 Xeon Phi 칩 상에서 AC 알고리즘을 병렬화한다. 이를 위하여 AC 알고리즘에서 입력 데이터에 대하여 여러 개의 패턴들을 동시에 매칭시키는 데에 사용되는 Deterministic Finite Automaton 구조를 압축시키는 새로운 기법을 제안한다. 이 기법은 캐시 미스를 감소시켜서 XeonPhi 상에서 AC 알고리즘의 성능을 크게 향상시킨다.