• Computers and Concrete
• /
• 제2권5호
• /
• pp.345-366
• /
• 2005

This work presents an assessment of the computational performance of a vector-parallel implementation of probabilistic model for concrete cracking in 3D. This paper shows the continuing efforts towards code optimization as reported in earlier works Paz, et al. (2002a,b and 2003). The probabilistic crack approach is based on the direct Monte Carlo method. Cracking is accounted by means of 3D interface elements. This approach considers that all nonlinearities are restricted to interface elements modeling cracks. The heterogeneity governs the overall cracking behavior and related size effects on concrete fracture. Computational kernels in the implementation are the inexact Newton iterative driver to solve the non-linear problem and a preconditioned conjugate gradient (PCG) driver to solve linearized equations, using an element by element (EBE) strategy to compute matrix-vector products. In particular the paper analyzes code behavior using OpenMP directives in parallel vector processors (PVP), such as the CRAY SV1 and CRAY T94. The impact of the memory architecture on code performance, and also some strategies devised to circumvent this issue are addressed by numerical experiment.

• 대한기계학회논문집A
• /
• 제22권1호
• /
• pp.132-141
• /
• 1998

A parallel version of sheet forming analysis program was developed. This version is compatible with any parallel computers which support MPI that is one of the most recent and popular message passing libraries. For this purpose, SERI-SFA, a vector version which runs on Cray Y-MP C90, a sequential vector computer, was used as a source code. For the sake of the effectiveness of the work, the parallelization was focused on the selected part after checking the rank of CPU consumed from the exemplary calculation on Cray Y-MP C90. The subroutines associated with contact algorithm was selected as targe parts. For this work, MPI was used as a message passing library. For the performance verification, an oil pan and an S-rail forming simulation were carried out. The performance check was carried out by the kernel and total CPU time along with theoretical performance using Amdahl's Law. The results showed some performance improvement within the limit of the selective paralellization.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 하계학술대회 논문집 D
• /
• pp.2281-2283
• /
• 2004

In this paper, we show the vector control performances for the parallel-connected motor drive system using the indirect vector control and the proposed vector control. The suggested estimation scheme of the rotor flux position is presented to reduce the sensitivity due to the load difference between the motors. To confirm the validity of the proposed control method, we compare the simulation results of the proposed control method with those of the conventional indirect vector control method. The simulation results show that the proposed control method is more effective for a change in the load torque.

• Journal of Positioning, Navigation, and Timing
• /
• 제1권1호
• /
• pp.1-6
• /
• 2012

For a Global Positioning System (GPS) receiver, it is known that a Vector Delay Locked Loop (DLL) in which the code signals of each satellite are tracked in parallel by using navigation results shows better performance in the aspect of the tracking accuracy and the robustness than that of a Scalar DLL. However, the quantitative analysis and the logical grounds for that performance enhancement of the Vector DLL are not sufficient. This paper, therefore, proposes the structure of the GPS receiver with the Vector DLL and analyzes the performance of it. The tracking and the positioning accuracy of the Vector DLL are theoretically analyzed and confirmed by simulation results. From the simulation results, it can be seen that the tracking and positioning accuracy has been improved about 30% in case that the receiver is static and the positioning is conducted for every Pre-detection Integration Time (PIT) while C/N0 is 45 dB-Hz.

• 정보처리학회논문지A
• /
• 제16A권6호
• /
• pp.411-418
• /
• 2009

퍼지 클러스터링 기반 벡터 양자화 알고리즘은 퍼지 클러스터링 분석이 벡터 양자화 프로세스 초기단계에서 초기화에 덜 민감하게 하기 때 문에 데이터 압축 분야에서 널리 사용되어 왔다. 하지만, 퍼지 클러스터링 처리는 훈련 벡터 공간에 포함된 불확실한 양적 공식의 복잡한 프레 임워크 때문에 상당한 계산량이 요구된다. 이러한 상당한 계산량 부하를 극복하기위해 본 논문은 4,096 프로세싱 엘리먼트로 구성된 어레이 아 키텍처를 이용하여 퍼지 벡터 양자화 알고리즘의 병렬 구현을 제안한다. 제안하는 병렬 구현은 4,096 프로세싱 엘리먼트를 이용하여 클러스터 링 프로세스 동안 효과적인 벡터 할당 정책을 적용함으로써 계산적으로 효율적인 솔루션을 제공한다. 모의실험 결과, 제안한 병렬 구현은 기존 의 다른 어레이 아키텍처를 이용한 구현보다 성능 및 효율 측면에서 상당한 향상을 보였다. 또한동일한 130nm 기술에서 제안한 병렬 구현은 오늘날의 ARM이나 TI DSP 프로세서를 이용한 구현과 비교하여 약 1000배의 성능 향상 및 100배의 에너지 효율 향상을 보였다. 이 결과들은 향상된 성능 및 에너지효율에서 제안한 병렬 구현의 잠재가능성을 입증한다.

• Journal of Power Electronics
• /
• 제10권6호
• /
• pp.686-693
• /
• 2010

In this paper, a fault diagnosis method based on fuzzy logic for the three-parallel power converter in a wind turbine system is presented. The method can not only detect both open and short faults but can also identify faulty switching devices without additional voltage sensors or an analysis modeling of the system. The location of a faulty switch can be indicated by six-patterns of a stator current vector and the fault switching device detection is achieved by analyzing the current vector. A fault tolerant algorithm is also presented to maintain proper performance under faulty conditions. The reliability of the proposed fault detection technique has been proven by simulations and experiments with a 10kW simulator.

• Journal of Advanced Marine Engineering and Technology
• /
• 제39권2호
• /
• pp.172-178
• /
• 2015

Constructing a bilingual multi-word lexicon is confronted with many difficulties such as an absence of a commonly accepted gold-standard dataset. Besides, in fact, there is no everybody's definition of what a multi-word unit is. In considering these problems, this paper evaluates and analyzes the context vector approach which is one of a novel alignment method of constructing bilingual lexicons from parallel corpora, by comparing with one of general methods. The approach builds context vectors for both source and target single-word units from two parallel corpora. To adapt the approach to multi-word units, we identify all multi-word candidates (namely noun phrases in this work) first, and then concatenate them into single-word units. As a result, therefore, we can use the context vector approach to satisfy our need for multi-word units. In our experimental results, the context vector approach has shown stronger performance over the other approach. The contribution of the paper is analyzing the various types of errors for the experimental results. For the future works, we will study the similarity measure that not only covers a multi-word unit itself but also covers its constituents.

• 산업경영시스템학회지
• /
• 제38권3호
• /
• pp.56-63
• /
• 2015

Several fields of science have demanded large-scale workflow support, which requires thousands of CPU cores or more. In order to support such large-scale scientific workflows, high capacity parallel systems such as supercomputers are widely used. In order to increase the utilization of these systems, most schedulers use backfilling policy: Small jobs are moved ahead to fill in holes in the schedule when large jobs do not delay. Since an estimate of the runtime is necessary for backfilling, most parallel systems use user's estimated runtime. However, it is found to be extremely inaccurate because users overestimate their jobs. Therefore, in this paper, we propose a novel system for the runtime prediction based on workload-aware clustering with the goal of improving prediction performance. The proposed method for runtime prediction of parallel applications consists of three main phases. First, a feature selection based on factor analysis is performed to identify important input features. Then, it performs a clustering analysis of history data based on self-organizing map which is followed by hierarchical clustering for finding the clustering boundaries from the weight vectors. Finally, prediction models are constructed using support vector regression with the clustered workload data. Multiple prediction models for each clustered data pattern can reduce the error rate compared with a single model for the whole data pattern. In the experiments, we use workload logs on parallel systems (i.e., iPSC, LANL-CM5, SDSC-Par95, SDSC-Par96, and CTC-SP2) to evaluate the effectiveness of our approach. Comparing with other techniques, experimental results show that the proposed method improves the accuracy up to 69.08%.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제1권1호
• /
• pp.35-53
• /
• 1997

This paper analyzes the time complexity of Genetic Algorithm based Task Scheduling (GATS) which is designed for the scheduling of parallel programs with diverse embedded parallelism types in a heterogeneous network systems. The analysis of time complexity is performed based on two representation methods (REIA, REIS) which are proposed in this paper to encode the scheduling information. And the heterogeneous network systems consist of a set of loosely coupled parallel and vector machines connected via a high-speed network. The objective of heterogeneous network computing is to solve computationally intensive problems that have several types of parallelism, on a suite of high performance and parallel machines in a manner that best utilizes the capabilities of each machine. Therefore, when scheduling in heterogeneous network systems, the matching of the parallelism characteristics between tasks and parallel machines should be carefully handled in order to obtain more speedup. This paper shows how the parallelism type matching affects the time complexity of GATS.

• 정보처리학회논문지D
• /
• 제10D권3호
• /
• pp.459-470
• /
• 2003

데이터웨어하우징의 애트리뷰트 벡터나 멀티미디어 데이터베이스의 특징 벡터는 모두 고차원 데이터를 이루고 있기 때문에, 이러한 고차원 데이터를 효율적으로 검색하기 위해서는 고차원 색인 기법이 요구된다. 이를 위하여 다수의 고차원 색인 기법들이 제안되었는데, 제안된 대부분의 색인 기법들이 차원의 수가 증가할수록 검색 성능이 급격히 저하되는 ‘차원 저주(dimensional curse)’ 문제를 지니고 있다. 셀-기반 필터링(Cell-Based Filtering : CBF) 기법은 이러한 차원 저주 문제를 해결하기 위해 제안되었다. 그러나 CBF 기법은 데이터의 양이 증가할수록 선형적으로 검색 성능이 감소하며, 이를 극복하기 위해 병렬 처리 기법을 사용하는 것이 필요하다. 본 논문에서는 데이터 디클러스터링(declustering) 방법으로 수평 분할 방식을 사용한 병렬 CBF 기법을 제안한다. 아울러 제안한 병렬 CBF 기법의 성능을 최대화하기 위하여, 병렬 CBF 기법을 다수의 서버로 구성된 Shared Nothing(SN) 구조의 클러스터 아키텍쳐 하에서 구축한다. 또한 SN 구조의 클러스터 아키텍쳐에 적합한 데이타 삽입 알고리즘, 범위질의 처리 알고리즘, k-최근접 질의 처리 알고리즘을 제시한다. 마지막으로 제안하는 병렬 CBF 기법이 기존 CBF 기법과 비교하여 서버 개수에 비례하여 우수한 검색 성능을 달성함을 보인다.