• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권9호
• /
• pp.3539-3558
• /
• 2015

This paper investigates and analyzes the default block placement strategy of HDFS. HDFS is a typical representative distributed file system to stream vast amount of data effectively at high bandwidth to user applications. However, the default HDFS block placement policy assumes that all nodes in the cluster are homogeneous, and places blocks with a simple RoundRobin strategy without considering any nodes' resource characteristics, which decreases self-adaptability of the system. The primary contribution of this paper is the proposition of a network load sensitive block placement strategy. We have implemented our algorithm and justify it through extensive simulations and comparison with similar existing studies. The results indicate that our work not only performs much better in the data distribution but also improves write performance more significantly than the others.

• 응용통계연구
• /
• 제30권6호
• /
• pp.931-941
• /
• 2017

반복이 있는 랜덤화 블록 계획법(randomized block design with replications)에서의 대표적인 검정법은 Mack이 제안한 방법과 Mack과 Skillings이 제안한 방법이 있다. 이 방법은 각 블록의 처리에서 반복된 각 관측값 대신에 반복된 관측값들의 평균을 이용하여 순위를 매기기 때문에 정보의 손실이 발생할 가능성이 있다. 이를 보완하기 위해 본 논문에서는 Chung과 Kim (2007)이 제안한 결합위치(joint placement) 방법에 점수함수(score function)를 적용한 선형위치통계량(linear placement statistics)을 이용한 검정방법을 제안하였다. 또한 Monte Carlo simulation study를 통해 기존의 방법들과 검정력을 비교하였다.

• 구강회복응용과학지
• /
• 제22권2호
• /
• pp.161-171
• /
• 2006

Endosseous implants have restored normal function and dental health to many patients. When implants were introduced as an effective treatment modality, their efficacy was limited by the amount of available bone. Today, various grafting procedures can surgically create bone width and volume. Implants can be placed in more ideal locations for successful prosthetic reconstruction. The use of autogenous bone grafts represents the "gold standard" for bone augmentation procedures. Either intraoral or extraoral sites may be considered for donor sites. Alveolar ridge augmentation using autogenous bone block, can be done during implant placement or staged with implant placement, after bone graft healing. In the staged technique, a better implant positioning and the use of wide diameter implants are possible. Alveolar ridge augmentation using autogenous block graft is a predictable way of treatment, for the atrophic alveolar ridge before implant placement. The cases presented in this article clinically demonstrate the efficacy of using a autogenous block graft in generating effective new bone fill for dental implant placement.

• 콘크리트학회논문집
• /
• 제11권3호
• /
• pp.59-67
• /
• 1999

The mass concrete structures are generally constructed in an incremental manner by deviding the whole structures by a series of many blocks. The temperature and stress distributions of any specific block are continuously affected by the blocks placed before and after the specific block. For an accurate analysis of mass concrete structures, the sequence of all the blocks must be accordingly considered including the change of material properties with time for those blocks considered. The purpose of this study is to propose a realistic analysis method which can take into account not only the influence of the sequence, time interval and size of concrete block placement on the temperatures and stresses, but also the change of material properties with time. It is seen from this study that the conventional simplified analysis, which neglects material property changes of some blocks with time and does not consider the effect of adjacent blocks in the analysis, may yield large discrepancies in the temperature and stress distributions of mass concrete structures. This study gives a method to choose the minimum number of blocks required to obtain reasonably accurate results in analysis. The study provides a realistic method which can determine the appropriate size and time interval of block placement, and can be efficiently used in the design and construction of mass concrete structures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권9호
• /
• pp.4063-4086
• /
• 2016

Cloud computing is a robust technology, which facilitate to resolve many parallel distributed computing issues in the modern Big Data environment. Hadoop is an ecosystem, which process large data-sets in distributed computing environment. The HDFS is a filesystem of Hadoop, which process data blocks to the cluster nodes. The data block placement has become a bottleneck to overall performance in a Hadoop cluster. The current placement policy assumes that, all Datanodes have equal computing capacity to process data blocks. This computing capacity includes availability of same storage media and same processing performances of a node. As a result, Hadoop cluster performance gets effected with unbalanced workloads, inefficient storage-tier, network traffic congestion and HDFS integrity issues. This paper proposes a storage-tier-aware Robust Data Placement (RDP) scheme, which systematically resolves unbalanced workloads, reduces network congestion to an optimal state, utilizes storage-tier in a useful manner and minimizes the HDFS integrity issues. The experimental results show that the proposed approach reduced unbalanced workload issue to 72%. Moreover, the presented approach resolve storage-tier compatibility problem to 81% by predicting storage for block jobs and improved overall data block placement by 78% through pre-calculated computing capacity allocations and execution of map files over respective Namenode and Datanodes.

• Journal of the Korean Data and Information Science Society
• /
• 제22권6호
• /
• pp.1105-1112
• /
• 2011

반복이 있는 랜덤화 블록 계획법을 검정하는 검정법은 Mack (1981), Skilling과 Wolfe (1977, 1978), Hettmansperger (1975)가 제안한 방법이 있다. 본 논문에서는 Orban과 Wolfe (1982)가 제안한 위치와 Kim (1999)이 제안한 대조군과 처리군의 방법을 확장하여 반복이 있는 랜덤화 블록 모형에서 새로운 방법을 제안하였다. 또한 모의 실험을 통하여 다른 비모수 검정 방법들과 모수 검정 방법의 검정력을 비교하였다.

• 전자공학회논문지A
• /
• 제33A권11호
• /
• pp.128-139
• /
• 1996

In this paper, we propose an algorithm to solve the problem of placement of rectangular blocks whose sizes and shpaes are pre-determined. The proposed method solves the placement of many retangular blocks of different sizes and shapes in a hierarchical manner, so as to minimize the chip area. The placement problem is divided into several sub-problems: hierarchical partioning, hierarchical area/shape estimation, hierarchical pattern pacement, overlap removal, and module rotation. After the circuit is recursively partitioned to build a hierarchy tree, the necessary wiring area and module shpaes are estimated using the resutls of the partitioning and the pin information before the placement is performed. The placement templaes are defined to represent the relative positions of the modules. The area and the connectivity are optimized separately at each level of hierachy using the placement templates, so the minimization of chip area and wire length can be achieved in a short execution time. Experiments are made on the MCNC building block benchmark circuits and the results are compared with those of other published methods. The proposed technique is shown to produce good figures in tems of execution time and chip area.

• 응용통계연구
• /
• 제29권7호
• /
• pp.1411-1419
• /
• 2016

랜덤화 블록 계획법을 검정하는 비모수적 방법에는 일반적인 대립가설에서 Friedman (1937)이 제안한 검정법이 있다. 이 방법은 처리 효과의 차이를 알아보기 위한 검정법으로 블록 내 순위를 사용해 검정함으로써 블록 간 정보의 손실이 있을 수 있다. 본 논문에서는 Hodges와 Lehmann (1962)이 제안한 정렬방법을 이용하여 블록 간 정보 손실을 줄이고, Jo와 Kim (2013)이 제안한 랜덤화 블록 계획법의 결합위치 방법을 확장하여 결합위치에 점수함수를 적용한 새로운 비모수적 방법을 제시하였다. 또한 Monte carlo simulation을 통하여 기존의 검정 방법과 제안한 검정법의 검정력을 비교하였다.

• 대한전자공학회논문지
• /
• 제26권2호
• /
• pp.147-154
• /
• 1989

Custom VLSI 칩의 설계에 있어서 임의의 폭과 높이를 갖는 직사각형 마크로 블럭을 효과적으로 배치하는 것은 칩의 면적과 신호 지연시간을 줄이기 위하여 매우 중요하다. 이 논문에서는 시뮬레이티드 어닐링을 사용하여 마크로 블럭을 전체적으로 최적 배치하기 위한 방법과, 시뮬레이티드 어닐링 과정이 끝난 후에 남아있는 직사각형 블럭들 사이의 중첩을 제거하는 효율적인 알고리듬을 제안하였다. 최소한의 배선영역을 확보하기 위해 각 블럭을 4방향으로 확장한 후에, 확장된 블럭들을 최대한 밀집되도록 배치하였다. 이 방법을 MV10000/UNIX 컴퓨터에서 C언어로 프로그램 하였으며 50개와 160개의 블럭으로 구성된 회로에 적용한 결과 좋은 배치를 얻을 수 있었다. 또한 최종 배치에 큰영향을 주는 파라미터에 대한 조사를 하였다.

• 응용통계연구
• /
• 제30권2호
• /
• pp.291-299
• /
• 2017

반복이 있는 랜덤화 블록 계획법을 검정하는 비모수 검정방법에는 Mack과 Skillings (1980), Mack (1981)가 제안한 방법이 있다. 본 논문에서는 Hodges와 Lehmann (1962)의 정렬 방법과 Chung과 Kim (2007)이 제안한 결합위치 검정법을 확장하여 반복이 있는 랜덤화 블록 모형에서 새로운 비모수적 방법을 제시하였다. 또한 모의실험을 통해 모수적 방법과 기존의 비모수적 방법과의 검정력을 비교하였다.