• The KIPS Transactions:PartA
• /
• v.11A no.7 s.91
• /
• pp.499-510
• /
• 2004

This paper proposes a heuristic load balancing algorithm for multiprocessor systems. The algorithm minimizes the number of idle links to distribute load traffic and reduces its communication cost. Each processor iteratively tries to transfer unit load to/from every neighbor processors. However, real load transfer is collectively done after complete load traffic calculation to minimize useless traffic. The proposed algorithm can be employed in various interconnection topologies with slight modifications. In this paper, it is applied to both hypercube and mesh environments. For performance evaluation, simulation studies are performed. The performance of proposed algorithm is compared to those of two well-known algorithms. The results show that the proposed algorithm always balances the loads perfectly. Furthermore, it reduces the communication costs by $70{\%}{\~}90{\%}$ in the hypercube ; and it reduces the cost by $\75{\%}$ in the mesh, compared to existing algorithms.

• KIISE Transactions on Computing Practices
• /
• v.21 no.8
• /
• pp.537-542
• /
• 2015

Database Shard is a technique that can be queried and stored by dividing one logical table into multiple databases horizontally. In order to analyze the shard data with aggregate or analysis functions, a process is required that integrates partial results on each shard database. In this paper, we introduce the design and implementation of a distributed processing system for aggregation and analysis on the CUBRID Shard distributed database, which is an open source database management system. The implemented system can accelerate the analysis onto multiple shards of partitioned tables; it shows efficient aggregation on shard distributed databases compared to stand-alone databases.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.19 no.6
• /
• pp.72-79
• /
• 2015

An efficient and practical reliability evaluation method is proposed for the coastal structures in this paper. It is capable of evaluating reliability of real complicated coastal structures considering uncertainties in various sources of design parameters, such as wave and current loads, resistance-related design variables including Young's modulus and compressive strength of the reinforced concrete, soil parameters, and boundary conditions. It is developed by intelligently integrating the Latin Hypercube sampling (LHS), Monte Carlo simulation (MCS) and the finite element method (FEM). The LHS-based MCS is used to significantly reduce the computational effort by limiting the number of simulation cycles required for the reliability evaluation. The applicability and efficiency of the proposed method were verified using a caisson-type breakwater structure in the numerical example.

• The Magazine of the IEIE
• /
• v.20 no.7
• /
• pp.104-114
• /
• 1993

본 연구 개발에서는 계속적으로 그 필요가 증가하는 대규모 계산을 위한 슈퍼급의 성능을 가지는 병렬처리 시스템의 구현, 구축을 목표로 하고 있다. 우리는 슈퍼급의 컴퓨터 시스템을 개발하기 위해 최신의 i860 프로세서를 사용한 단위컴퓨터를 설계, 제작하여 최대 40 MFLOPB의 성능을 가지는 단위 컴퓨터를 구현하였고, 8개의 단위컴퓨터를 연결하여 320MFLOPS컴퓨터를 개발하였다. 최종 목표는 128개의 단위 컴퓨터를 연결하여 5 GFLOPS 최대 계산능력을 갖는 7차원 하이퍼큐브 컴퓨터 시스템 KAICUBE-860을 구성하고자 한다. 대부분의 사람들은 그 생각방식이나 행동형태가 순차적인 구조를 가지고 있기 때문에 병렬적인 프로그램의 개발이 쉽지가 않고 때문에 사용자로 하여금 보다 쉽게 병렬시스템에 접근할 수 있도록 사용자 환경을 적절히 구축하여야만 한다. 따라서 우리의 최종 목표는 슈퍼급 성능의 병렬 처리 시스템이고 그 사용 방법은 보통순차적 컴퓨터를 사용하듯이 간단한 사용자 환경을 제공하는 시스템이다. 이를 위해 시스템 소프트웨어와 사용자 환경을 계속 개발 중에 있다.

• Journal of the Korea Society of Computer and Information
• /
• v.4 no.4
• /
• pp.69-76
• /
• 1999

Recently, it has been studied the methods to materialize and precompute the query results for complexed spatial aggregation queries with high response time and the popular use in spatial data warehouse. In this paper, we propose extended selective materialization algorithm and present the way to materialize selectively which is considered access frequency and computation time of spatial operation according to spatial measures of spatial views for improvement of existing selective materialization algorithms.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.3
• /
• pp.236-242
• /
• 2004

In this paper, we propose an approach to parallelize a load balancing algorithm that was shown to be very effective in distributing workload for parallel computations. Load balancing algorithms are required in executing parallel program efficiently As a parallel computation model, we used dynamically growing tree structure that can be found in many application problems. The load balancing algorithm tries to balance the workload among processors while keeping the communication cost under certain limit. We show how the load balancing algorithm is effectively parallelized on mesh and hypercube interconnection networks, and analyzed the time complexity for each case to show that parallel algorithm actually reduced the various overhead.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.2
• /
• pp.186-200
• /
• 2010

In order to render global illumination realistically, we need to accurately compute the direct and indirect illumination that represents the light information incoming through complex light paths. In this process, the indirect illumination at given point is greatly affected by surrounding geometries. Harmonic mean distance is a mathematical tool which is often used as a metric indicating the distance from a surface point to its visible objects in 3D space, and plays a key role in such advanced global illumination algorithms as irradiance/radiance caching and ambient occlusion. In this paper, we analyze the accuracy of harmonic mean distance estimated against various environments in the final gathering and photon mapping methods. Based on the experimental results, we discuss our experiences and future directions that may help develop an effective harmonic mean distance computation method in the future.

• Journal of KIISE:Databases
• /
• v.27 no.2
• /
• pp.165-174
• /
• 2000

Recent researches about range queries in OLAP are only concerned with applying an aggregation operator over a certain region. However, data analysts in real world need not only the simple range query pattern but also an extended range query pattern that finds ranges which satisfy a special condition specified by using several aggregation operators. In this work, we define the general form of the extended range query and propose an efficient processing method for the 'MAX -of-SUM' query, which is the representative form of the extended range query pattern. The MAX-of-SUM query finds the range which has the maximum range sum value in data cube where the size of the range is given. The proposed query processing method is based on the prediction of the scope of the range sum values. That is, the search space on the query processing can be reduced by using the result of the prediction, and hence, the query response time is also reduced.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.1
• /
• pp.35-44
• /
• 2005

This paper proposes an efficient heuristic load balancing algorithm for multiprocessor systems. The algorithm minimizes the number of idle links to distribute load traffic and reduces its communication cost. Each processor iteratively tries to transfer unit load to/from all neighbor processors. However, real load transfer is collectively done after all load traffic is calculated. This prevents useless traffic and thus reduces the overall load traffic. The proposed algorithm can be employed in various interconnection topologies with slight modifications. In this paper, it is applied to hypercube, mesh, k-ary n-cube and general graph environments. For performance evaluation, simulation studies are performed. The proposed algorithm and the well-known existing algorithms are implemented and compared. The results show that the proposed algorithm always balances the loads perfectly. furthermore, in comparison with the existing algorithms, it reduces the communication costs by 77%, 74% and 73% in the hypercube, the mesh, and k-ary n-cube, respectively.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10b
• /
• pp.118-120
• /
• 2004

OLAP의 주요 목표는 대용량 데이터를 다차원적으로 분석하여 그 결과를 신속하게 제공함으로써, 사용자의 의사 결정을 지원하는 것이다. 다차원 분석을 용이하게 하기 위해 차원 계층이 사전에 정의되고 표준화된 연산들이 사용되는데, 그러한 연산들로는 롤업, 드릴다운, 슬라이스, 다이스, 피봇을 들 수 있다. 그러나 분석과정에서 기존에 정의된 데이터의 그룹핑 이외의 다른 그룹핑 방식이 필요할 때가 있으며, 그러한 그룹핑으로 전체 데이터를 분석하는 것을 트렌드 분석이라고 한다. 트렌드 분석은 기업의 의사 결정에 매우 중요한 요소이나, 사전에 정의된 계층을 사용하는 것이 아니므로, 질의 처리 시에 트렌드 분석을 신속하게 하기는 어렵다. 본 연구에서는 트렌드 분석을 고속화하기 위한 방안으로 동적 계층 구조 생성을 제안한다. 특정 차원 기준으로 특정 범위 값으로 합산된 범위 합을 구하기 위해, 기존에 연구되었던 프리픽스섬 방식을 분석하여 문제점을 제시하였고 새로운 기법을 제안하였다. 또한 분석 시에 디스크 접근을 효율적으로 하기 위한 큐브 저장 방식을 제안하였다. 본 연구에서 제안한 방식으로 트렌드 분석을 하였을 때 접근해야 하는 디스크 블록 수도 계산하여 제안한 방법의 효율성을 검증하였다.