• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.32 no.2
• /
• pp.117-124
• /
• 2019

Parallel sparse solvers are essential for solving large-scale finite element models. This paper introduces the combination of iterative and direct solver that can be applied efficiently to problems that require continuous solution for a subtly changing sequence of systems of equations. The iterative-direct sparse solver combination technique, proposed and implemented in the parallel sparse solver package, PARDISO, means that iterative sparse solver is applied for the newly updated linear system, but it uses the direct sparse solver's factorization of previous system matrix as a preconditioner. If the solution does not converge until the preset iterations, the solution will be sought by the direct sparse solver, and the last factorization results will be used as a preconditioner for subsequent updated system of equations. In this study, an improved method that sets the maximum number of iterations dynamically at the first Krylov iteration step is proposed and verified thereby enhancing calculation efficiency by the frequency domain analysis.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.11a
• /
• pp.202-203
• /
• 2017

비음수 행렬 인수분해(Non-negative Matrix Factorization, NMF)의 신호분리 성능을 개선하기 위해 희소조건을 인가한 방법이 희소 비음수 행렬 인수분해 알고리즘(Sparse NMF, SNMF)이다. 기존의 SNMF 알고리즘은 개별 음원의 희소성을 고려하지 않고 임의로 결정한 희소 조건을 사용한다. 본 논문에서는 음원의 특성에 따른 희소성을 추정하고 이를 SNMF 학습알고리즘에 적용하는 새로운 신호분리 기법을 제안한다. 혼합 신호에서의 잡음제거 실험을 통해, 제안한 방법이 기존의 NMF와 SNMF에 비해 성능이 더 우수함을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.45-47
• /
• 2021

APU(Accelerated Processing Unit)는 CPU와 GPU가 통합되어있는 프로세서이며 같은 메모리 공간을 사용한다. CPU와 GPU가 분리되어있는 기존 이종 컴퓨팅 환경에서는 GPU가 작업을 처리하기 위해 CPU에서 GPU로 메모리 복사가 이루어졌지만, APU는 같은 메모리 공간을 사용하므로 메모리 복사 없이 가상주소 할당으로 같은 물리 주소에 접근할 수 있으며 이를 Zero Copy라 한다. Zero Copy 성능을 테스트하기 위해 희소행렬 연산을 사용하였으며 기존 메모리 복사대비 크기가 큰 데이터는 약 4.67배, 크기가 작은 데이터는 약 6.27배 빨랐다.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.5
• /
• pp.409-416
• /
• 2017

In this paper, we consider a framework of compressed sensing over finite fields. One measurement sample is obtained by an inner product of a row of a sensing matrix and a sparse signal vector. A recovery algorithm proposed in this study for sparse signals based probabilistic decoding is used to find a solution of compressed sensing. Until now compressed sensing theory has dealt with real-valued or complex-valued systems, but for the processing of the original real or complex signals, the loss of the information occurs from the discretization. The motivation of this work can be found in efforts to solve inverse problems for discrete signals. The framework proposed in this paper uses a parity-check matrix of low-density parity-check (LDPC) codes developed in coding theory as a sensing matrix. We develop a stochastic algorithm to reconstruct sparse signals over finite field. Unlike LDPC decoding, which is published in existing coding theory, we design an iterative algorithm using probability distribution of sparse signals. Through the proposed recovery algorithm, we achieve better reconstruction performance as the size of finite fields increases. Since the sensing matrix of compressed sensing shows good performance even in the low density matrix such as the parity-check matrix, it is expected to be actively used in applications considering discrete signals.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.4C
• /
• pp.440-446
• /
• 2007

This paper addresses a new representation of DFT matrix via the Jacket transform based on the element inverse processing. We simply represent the inverse of the DFT matrix following on the factorization way of the Jacket transform, and the results show that the inverse of DFT matrix is only simply related to its sparse matrix and the permutations. The decomposed DFT matrix via Jacket matrix has a strong geometric structure that exhibits a block modulating property. This means that the DFT matrix decomposed via the Jacket matrix can be interpreted as a block modulating process.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.3
• /
• pp.141-149
• /
• 2006

In this paper, we introduce an improved algorithm for computing matrix triple product that commonly arises in primal-dual optimization method. In computing $P=AHA^{t}$, we devise a single pass algorithm that exploits the block diagonal structure of the matrix H. This one-phase scheme requires fewer floating point operations and roughly half the memory of the generic two-phase algorithm, where the product is computed in two steps, computing first $Q=HA^{t}$ and then P=AQ. The one-phase scheme achieved speed-up of 2.04 on Intel Itanium II platform over the two-phase scheme. Based on memory latency and modeled cache miss rates, the performance improvement was evaluated through performance modeling. Our research has impact on performance tuning study of complex sparse matrix operations, while most of the previous work focused on performance tuning of basic operations.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10a
• /
• pp.537-541
• /
• 2006

최근 특정 주제의 지적 구조를 파악하기 위한 저자 동시인용분석, 동시단어분석, 서지결합법 등 계량정보분석에 대한 연구가 활발히 진행되고 있다. 그러나 국내의 경우 계량정보분석 기법을 활용한 정보 유통 프레임웍을 갖추고 있는 연구기관이나 대학이 아직 없는 실정이다. 그 이유는 특허나 과학문헌에 대한 인용정보를 보유한 곳이 없고, 거대 인용정보 행렬을 계산하기 위한 컴퓨팅 자원을 확보하지 못하고 있기 때문이다. 본 연구는 미국 특허 데이터베이스를 대상으로 인용 피인용 행렬을 생성한 후, 클러스터 컴퓨터를 사용하여 동시인용과 서지결합빈도를 계산하고 그 결과를 이용자에게 제공하는 정보 유통 서비스 모델을 제시하고자 한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.2
• /
• pp.205-211
• /
• 2012

The retrieved documents have to be transformed into proper data structure for the clustering algorithms of statistics and machine learning. A popular data structure for document clustering is document-term matrix. This matrix has the occurred frequency value of a term in each document. There is a sparsity problem in this matrix because most frequencies of the matrix are 0 values. This problem affects the clustering performance. The sparseness of document-term matrix decreases the performance of clustering result. So, this research uses the factor score by factor analysis to solve the sparsity problem in document clustering. The document-term matrix is transformed to document-factor score matrix using factor scores in this paper. Also, the document-factor score matrix is used as input data for document clustering. To compare the clustering performances between document-term matrix and document-factor score matrix, this research applies two typed matrices to self organizing map (SOM) clustering.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.30 no.2
• /
• pp.127-135
• /
• 2017

The computing environment has changed rapidly to enable large-scale finite element models to be analyzed at the PC or workstation level, such as multi-core CPU, optimal math kernel library implementing BLAS and LAPACK, and popularization of direct sparse solvers. In this paper, the design considerations on a parallel finite element code for shared memory based multi-core CPU system are proposed; (1) the use of optimized numerical libraries, (2) the use of latest direct sparse solvers, (3) parallelism using OpenMP for computing element stiffness matrices, and (4) assembly techniques using triplets, which is a type of sparse matrix storage. In addition, the parallelization effect is examined on the time-consuming works through a large scale finite element model.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.79-81
• /
• 2003

Kyonggi21Search시스템은 GIS와 웹을 통합한 지역정보 검색 시스템이다. 웹과 GIS를 연동하여 지리정보를 검색하기 위해 웹 문서에서 지역관련 색인어를 추출하고, 색인어간의 관련성을 계산한다. "Kyonggi21Search"시스템에서는 웹 문서에 많이 나타나는 일반적인 단어보다는, 많은 문서에 나타나지 않는 지리적 문화적인 단어들 간의 관련성을 찾는 것이 더 중요한데, 본 연구에서는 단어들 간의 관련성을 찾는데 연관규칙과 연관클러스터를 이용하여 연관도를 계산한다. 그리고 이런 단어들의 관련성을 찾는데는 연관 클러스터를 이용하는 것이 더 적합하다는 것을 보여준다. 한편 웹 문서와 색인어를 이용하여 만든 행렬은 희소행렬이라는 점을 이용하여 연관 클러스터 방법의 단점인 높은 계산량을 줄이는 최적화 방법을 제안한다.