• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2001년도 추계학술발표논문집 (상)
• /
• pp.313-316
• /
• 2001

At the end of 1999, the TeraCluster project in the KISTI Supercomputing Center was initiated to explore the possibility of PC clusters as a scientific computing platform to replace the Cray T3E system in KISTI by the year 2002. In order to understand whether an application is scalable to tera-flops sized cluster system, running test is inevitable. Extensive performance tests using well-known benchmarking codes with real applications' characteristics in them were carried out with different combinations of CPUs, system boards, network devices. The lessen learned shows the relationships between system performances and varied applications' different needs resulting in promises of How-Tos in building large scale cluster system. The 64/16 node clusters with Alpha EV6(466MHz), Pentium III(667 MHz) i inter-node network of Fast Ethernet. SCI[1] and Myrinet[2] were evaluated. More detailed specifications of the Linux clusters are described in Table 1.

• 천문학회보
• /
• 제36권1호
• /
• pp.58.2-58.2
• /
• 2011

Park, Gott, & Choi (2008) found that when a galaxy is located within the virial radius from its closest neighbor and the neighbor is an elliptical, the probability of the galaxy to be an elliptical is very sensitive to the large-scale background density over a few Mpc scales. They suggested that the large-scale dependence can be arise if the temperature of a diffuse hot gas held by elliptical galaxies are higher in higher density environment. In this study, to understand the large-scale environment affects the X-ray properties of individual galaxies, we investigated the dependence of the X-ray luminosities of the elliptical galaxies on the large-scale environment using X-ray and optical data which we selected from the ROSAT All-Sky Survey and the Sloan Digital Sky Survey Data Release 7. To exclude galaxies embedded in an intra-group/cluster medium which could enhance their observed X-ray luminosity, we used isolated elliptical galaxies.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제21권9호
• /
• pp.587-595
• /
• 2015

전통적으로 고에너지 물리, 해양, 기상, 천문 우주 등 다양한 과학 분야에서 수천 코어 이상의 CPU를 사용하는 대규모 워크플로우 지원을 요구하고 있으며 이를 위해 대부분 슈퍼컴퓨터와 같은 클러스터 기반의 대용량 시스템이 활용되고 있다. 이러한 시스템은 다수의 사용자 및 기관에 의해 공유되고 있으며, 사용자들의 다양한 요구 사항으로 인해 시스템 운영 및 관리에 많은 어려움이 있다. 본 논문에서는 가상화로 인한 성능 저하 문제를 최소화하고 사용자가 원하는 환경을 동적으로 제공하기 위해 컨테이너 기반 클러스터 관리 플랫폼 방안을 제시하고 구축 사례를 소개한다. 본 논문의 의의는 다음 3가지로 볼 수 있다. 먼저, 컨테이너 기반 가상화 기술과 스케줄러 기능을 연동하여 큰 성능 저하 없이 대규모의 과학워크플로우 지원을 위한 클러스터 구성 및 관리 방안을 제시하였다. 둘째, Docker 와 HTCondor를 활용하여 제시된 방안을 손쉽게 구축한 사례를 소개하였다. 셋째, 널리 활용되는 벤치마크 툴을 이용하여 Docker 성능을 검증하였으며, 다양한 프로그램 언어로 구현된 몬테카를로 시뮬레이션을 통해 과학 워크플로우 지원 예제를 제시하였다.

• Journal of the Korean Statistical Society
• /
• 제27권4호
• /
• pp.459-469
• /
• 1998

Balanced half sample method is a simple variance estimation method for complex sampling designs. Since it is simple and flexible, it has been widely used in large scale sample surveys. However, the usual BHS method overestimate the true variance in without replacement sampling and two-stage cluster sampling. Focusing on this point , we proposed an unbiased BHS variance estimator in a stratified two-stage cluster sampling and then described an implementation method of the proposed estimator. Finally, partially BHS design is explained as a tool of reducing the number of replications of the proposed estimator.

• 한국항공우주학회지
• /
• 제34권6호
• /
• pp.67-74
• /
• 2006

본 논문에서는 저가의 클러스터 컴퓨터 시스템과 저해상도 영상장비들을 이용하여 초대형 해석 데이터를 정밀하게 분석할 수 있는 고해상도 타일 가시화 시스템을 개발하였다. 타일 가시화 하드웨어 구축 시 유의점을 고찰하고, 화면왜곡 현상을 제거할 수 있는 빔프로젝터 위치조절장치를 설계/제작하였다. 타일 가시화 소프트웨어 개발에서 그래픽 사용자 인터페이스와 렌더링을 위해서는 Qt와 OpenGL 라이브러리를 이용하였다. 또한 LAM-MPI 라이브러리를 통해 각각의 클러스터 컴퓨터 노드로부터 얻게 되는 조각적인 화면들을 전체의 한 화면으로 동기화시켜 왜곡 없는 전체 타일 영상을 만들도록 하였다.

• 한국항공우주학회지
• /
• 제34권8호
• /
• pp.33-40
• /
• 2006

비선형 외연 유한요소법에서 유한요소 병렬 처리 방안을 기술하고 코드에 구현하였다. 성능테스트 장비로 자체 구축한 520 개의 CPU를 갖는 리눅스 클러스터 슈퍼컴퓨터를 사용하였다. 대규모 모델 테스트 결과 256 개의 CPU 까지도 거의 이상적인 속도 증가를 보였다. 유한요소 계산시간 대비 통신시간 계산이 전체 성능에 미치는 영향도 검토하였다. 사용 프로세서가 증가할수록 상용코드의 병렬 성능 대비 더 좋은 성능을 보이는 것으로 나타났다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1238-1259
• /
• 2019

Saving energy is a big challenge for Wireless Sensor Networks (WSNs), which becomes even more critical in large-scale WSNs. Most energy waste is communication related, such as collision, overhearing and idle listening, so the schedule-based access which can avoid these wastes is preferred for WSNs. On the other hand, clustering technique is considered as the most promising solution for topology management in WSNs. Hence, providing interference-free clustering is vital for WSNs, especially for large-scale WSNs. However, schedule management in cluster-based networks is never a trivial work, since it requires inter-cluster cooperation. In this paper, we propose a clustering method, called Interference-Free Clustering Protocol (IFCP), to partition a WSN into interference-free clusters, making timeslot management much easier to achieve. Moreover, we model the clustering problem as a multi-objective optimization issue and use non-dominated sorting genetic algorithm II to solve it. Our proposal is finally compared with two adaptive clustering methods, HEED-CSMA and HEED-BMA, demonstrating that it achieves the good performance in terms of delay, packet delivery ratio, and energy consumption.

• 한국CDE학회논문집
• /
• 제17권2호
• /
• pp.123-131
• /
• 2012

The requirements for computational resources to perform the structural analysis are increasing rapidly. The size of the current analysis problems that are required from practical industry is typically large-scale with more than millions degrees of freedom (DOFs). These large-scale analysis problems result in the requirements of high-performance analysis codes as well as hardware systems such as supercomputer systems or cluster systems. In this paper, the pre- and post-processing system for supercomputing based large-scale structural analysis is presented. The proposed system has 3-tier architecture and three main components; geometry viewer, pre-/post-processor and supercomputing manager. To analyze large-scale problems, the ADVENTURE solid solver was adopted as a general-purpose finite element solver and the supercomputer named 'tachyon' was adopted as a parallel computational platform. The problem solving performance and scalability of this structural analysis system is demonstrated by illustrative examples with different sizes of degrees of freedom.

• 천문학회보
• /
• 제46권2호
• /
• pp.41.3-42
• /
• 2021

Coma cluster (Abell 1656) is one of the most massive local galaxy clusters such as Virgo, Fornax, and Perseus, which holds a large collection of globular clusters. Globular cluster systems (GCSs) in a galaxy cluster tell us a history of hierarchical cluster assembly and intracluster GCs (ICGCs) are known to trace the gravitational potential of the galaxy cluster. Previous studies of GCSs in Coma mainly utilized data obtained using Hubble Space Telescope (HST) with high spatial resolution. However, most of the data were based on narrow-field pointing observations. In this study we present the widest survey of GCSs in the Coma cluster using the archival Subaru/Hyper Suprime-Cam (HSC) g and r images, supplemented with the archival HST images. The Coma GCSs are largely extended in E-W and SW direction, along the general direction of Coma-Abell 1367 filament. This global structure of the GCSs is consistent with the spatial distribution of the intracluster light (ICL). ICGC spatial distribution is largely extended to almost ~50% of the virial radius. Most of these ICGCs are blue and metal-poor, which supports the scenario that ICGCs are mainly originated from dwarf galaxies and some proportion from brighter galaxies. Implications of the results will be discussed.

• 천문학회지
• /
• 제37권5호
• /
• pp.477-482
• /
• 2004

Shock waves form in the intergalactic space as an ubiquitous consequence of cosmic structure formation. Using N-body/hydrodynamic simulation data of a ACDM universe, we examined the properties of cosmological shock waves including their morphological distribution. Adopting a diffusive shock acceleration model, we then calculated the amount of cosmic ray energy as well as that of gas thermal energy dissipated at the shocks. Finally, the dynamical consequence of those cosmic rays on cluster properties is discussed.