• 정보처리학회논문지A
• /
• 제14A권1호
• /
• pp.1-14
• /
• 2007

그리드 네트워크에서 MPI를 사용하여 지리적으로 산재된 컴퓨팅 자원을 활용하여 복잡한 문제를 해결하기 위한 MPICH-G2는 사설 IP 클러스터를 지원하지 못한다는 단점을 가지고 있다. MPICH-G2의 경우에는 모든 계산노드가 외부에 노출되어 악의적인 보안 침해의 가능성이 높아지게 된다. MPICH-GP는 NAT와 프락시를 병용하여 사설 IP 클러스터의 문제를 해결한다. 사설 IP 클러스터의 프론트 노드에 프락시를 두고, 이를 통해 내부 계산 노드로의 통신 링크를 중계한다. 따라서 사설 IP 클러스터와 공인 IP 클러스터가 혼재된 네트워크에서도 적정한 경로를 설정하고 성공적으로 MPI 작업을 수행할 수 있다. 본 논문에서는 MPICH-GP의 성능을 기존의 MPICH-G2와 비교하였다. 그리드 환경에서 MPICH-GP는 MPICH-G2의 80% 이상의 성능을 보이며, RANK 관리기법을 적용한 경우는 95% 이상의 성능을 나타낸다.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2012년도 제42회 동계 정기 학술대회 초록집
• /
• pp.571-571
• /
• 2012

We discovered the formation of C60 aggregates in solution by means of photoluminescence spectroscopic study on C60 in solutions. From the in-depth investigation of temperature dependence of the luminescence of C60 in toluene, benzene and CS2 solutions, we reported that the C60 aggregates are formed during cooling at the freezing temperature of these solvents. Furthermore, the C60 aggregates can be changed to stable structures by irradiating with UV pulse-laser (Nd:YAG laser, 355nm). As a consequence, we could obtain nano-scale photo-polymerized C60 clusters, which appear as round-shaped nano- scale particles in high resolution transmission electron-microscopy (HRTEM) images. However, the yield of the nano-scale C60 clusters obtained by this method is too small. So we designed and developed a system to obtain C60 cluster of macroscopic quantity by using ultrasonic nebulizer. In this system, C60 solution was vaporized to several micro-sized droplets in vacuum, resulting in the formation of C60 aggregates by evaporating solvent (toluene). The system was invented to produce nano-scale carbon clusters by the irradiation of UV light upon C60 aggregates in vacuum. We have characterized the products, C60 cluster, obtained from the system by using UV absorption spectra and HPLC spectra. Although the products have a possibility of inclusion various forms of C60 cluster, results support that the product formed from the system by using vaporizer method establishes a new method to obtain C60 cluster in macroscopic quantity. In the presentation, the details of the system and the results of characterization are reported.

• 보건교육건강증진학회지
• /
• 제30권2호
• /
• pp.23-39
• /
• 2013

Objectives: This study was conducted to investigate practical experiences of telemedicine of community health practitioners(CHPs). Methods: Qualitative data were collected by in-depth interviews from 10 CHPs who have experiences in managing telemedicine system. All interviews were recorded and transcribed according to qualitative conventional content analysis processes. Results: As a result, 32 themes were deduced and 11 theme clusters and 3 categories were formed and each coding categories were derived directly from the text data. 11 theme clusters derived from the 32 meaningful themes were as follows: Human resources, equipments and systems, computer program (Input resources), human resource management, patient registration and management, medication, laboratory test (Progress), benefits in telemedicine system managing, difficulties in telemedicine system managing, complains in telemedicine system managing, client responses to telemedicine system (Outcome evaluation). 3 categories derived 11 theme clusters were 'input', 'progress', and 'outcome evaluation'. Conclusions: This study has contributed to the understanding of operation of telemedicine by CHPs in community health posts. For more systematic and comprehensive management, further study should be conducted to reflect experience and positions of public health center physicians, collaborative hospital physician and patients.

• 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.

• 정보처리학회논문지B
• /
• 제14B권7호
• /
• pp.513-522
• /
• 2007

군집화는 주어진 데이터를 분할하여 데이터 속에 숨겨져 있는 의미를 자동으로 발견하는 방법으로, 사람이 일일이 살펴보기 어려운 데이터를 분석해서 비슷한 성향을 가진 데이터들끼리 모은 여러 개의 군집들을 만들어 낸다. 온라인 문서 군집화는 검색 엔진을 통해 검색된 문서들을 대상으로 군집화를 실행하여 유사한 특성의 문서들을 묶어서 보여줌으로써 사용자의 검색 환경의 편의성을 증진시키는 것이 목적이다. 문서군집화는 사람의 개입이 없이 자동으로 이루어져야 하고, 군집화 결과에 영향을 미치는 군집의 개수 선정도 자동으로 이루어져야 한다. 또한, 온라인 시스템에서는 빠른 응답 시간을 보장하는 것이 중요하다. 본 논문에서는 기하학적인 정보를 이용하여 군집의 수를 결정하는 방법을 제안한다. 제안하는 방법은 군집의 중심을 저차원 평면에 사상하는 것과 사상된 군집 중심의 거리 정보를 이용하여 군집들을 병합하는 두 단계로 이루어져 있다. 제안하는 방법을 실데이터에 적용하여 실험한 결과 군집화 성능이 향상되고, 처리 시간도 온라인 환경에 적합한 것을 확인 할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권7호
• /
• pp.3714-3732
• /
• 2019

With the rapid development of network, Intrusion Detection System(IDS) plays a more and more important role in network applications. Many data mining algorithms are used to build IDS. However, due to the advent of big data era, massive data are generated. When dealing with large-scale data sets, most data mining algorithms suffer from a high computational burden which makes IDS much less efficient. To build an efficient IDS over big data, we propose a classification algorithm based on data clustering and data reduction. In the training stage, the training data are divided into clusters with similar size by Mini Batch K-Means algorithm, meanwhile, the center of each cluster is used as its index. Then, we select representative instances for each cluster to perform the task of data reduction and use the clusters that consist of representative instances to build a K-Nearest Neighbor(KNN) detection model. In the detection stage, we sort clusters according to the distances between the test sample and cluster indexes, and obtain k nearest clusters where we find k nearest neighbors. Experimental results show that searching neighbors by cluster indexes reduces the computational complexity significantly, and classification with reduced data of representative instances not only improves the efficiency, but also maintains high accuracy.

• 천문학회보
• /
• 제45권1호
• /
• pp.32.2-32.2
• /
• 2020

Massive Compact Elliptical Galaxies (MCEGs) found in the local universe are as massive as normal galaxies but extremely compact (M∗ > 1011 Msun, Reff < 1.5 kpc). They are considered to be the relics of red nugget galaxies found at high redshift. They are not likely to have undergone many mergers, keeping their original mass and size. Moreover, it is expected that they host a dominant population of red (metal-rich) globular clusters rather than blue (metal-poor) ones. Indeed, Beasley et al. (2018) found that the color distribution of the cluster system of NGC 1277 is unimodal, showing only a red population. However, NGC 1277 is the only case whose cluster system was studied among MCEGs. In this study, we investigate globular cluster systems of 14 nearby MCEGs with a homogeneous data set of HST/WFC3 F814W/F160W archive images. We detect tens to hundreds of globular clusters in each galaxy and examine their color distributions. Surprisingly, the fractions of red globular clusters are similar to those of normal galaxies, and are much lower than that of NGC 1277. We additionally obtain Gemini/GMOS-N g'r'i' images of PGC 70520, one of the 14 nearby MCEGs, to detect more globular clusters from deeper and wider images. We will discuss the results from the Gemini data combined with the results from the HST data in relation with the formation of MCEGs.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 추계종합학술대회 논문집
• /
• pp.403-407
• /
• 1999

In this paper we present an architecture, implementation, and performance evaluation of an adaptive communication system (ACS) for wide-area ATM clusters. Our approach capitalizes on thread-based programming model to overlap computation and communication and develop a dynamic message-passing environment with separate data and control paths. This leads to a flexible and adaptive message-passing environment that can support multiple flow control, error control, and multicasting algorithms. We show the performance of ACS applications and compare them with p4, PVM, and MPI applications.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2004년도 추계학술대회
• /
• pp.221-229
• /
• 2004

Recently, there has been enormous growth in the amount of commercial and scientific data, such as protein sequences, retail transactions, and web-logs. Such datasets consist of sequence data that have an inherent sequential nature. However, only a few of the existing clustering algorithms consider sequentiality. This study presents a method for clustering such sequence datasets. The similarity between sequences must be decided before clustering the sequences. This study proposes a new similarity measure to compute the similarity between two sequences using a sequence element. Two clustering algorithms using the proposed similarity measure are proposed: a hierarchical clustering algorithm and a scalable clustering algorithm that uses sampling and a k-nearest neighbor method. Using a splice dataset and synthetic datasets, we show that the quality of clusters generated by our proposed clustering algorithms is better than that of clusters produced by traditional clustering algorithms.

• 디지털산업정보학회논문지
• /
• 제5권1호
• /
• pp.23-32
• /
• 2009

Although the web is becoming a widely accepted medium, it provides relatively poor performance and low availability. A cluster consists of a collection of interconnected stand-alone computers working together and provides a high-availability solution in application area such as web services or information systems. Web server clusters require a high-availability service with a proactive and practical fault management. However, as the system complexity grows, it is not easy to meet the requirement. Therefore, web server clusters must have self-fault management capability for meeting high-availability requirement. In this paper, we propose high availability web server clusters using self-healing technique with a minimal human intervention. Our experimental results show that a proposed method can be used to improve the availability of web server clusters.