• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.21 no.4
• /
• pp.317-324
• /
• 2008

Optimum design for general or complex structures are required to the need of many numbers of structural analyses. However, current computational environment with single processor is not capable of generating a high-level efficiency in structural analysis and design process for complex structures. In this paper, a virtual parallel computing system communicated by an internet of personal computers and workstation is constructed. In addition, a routine executing Pro/E, ANSYS and optimization algorithm automatically are adopted in the distributed process technique of sequential approximate optimization for the purpose of enhancing the flexibility of application to general structures. By employing the distributed processing technique during structural analysis using commercial application, total calculation time could be reduced, which will enhance the applicability of the proposed technique to the general complex structures.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.6
• /
• pp.556-565
• /
• 1991

Using the WDZ decomposition algorithm, a parallel algorithm is presented for solving the linear system Ax=b which has an nxn nonsingular tridiagonal matrix. For implementing this algorithm a CAM systolic arrary is proposed, and each processing element of this array has its own CAM to store the nonzero elements of the tridiagonal matrix. In order to evaluate this array the algorithm presented is compared to theis compared to the LU decomposition algorithm. It is found that the execution time of the algorithm presented is reduced to about 1/4 than that of the LU decomposition algorithm. If each computation process step can be dome in one time unit, the system of eqations is solved in a systolic fashion without central control is obtained in 2n+1 time steps.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.2 no.1
• /
• pp.87-99
• /
• 1998

In the multicasting communications, the quality of service is not guaranteed occasionally, according to the shortage of resources, because all resources such as communication buffer, processing power for the packet multiplication, and bandwidth, should be shared among each members. This problem can be solved through the hierarchical multicaster construction mechanism which guarantees that all multicaster can be operated within the required performance. This paper proposes the architecture of multipoint communication server based on hypercube type massively parallel computer. Through the simulation, it is verified that the traffic of links between each node of hypercube computer is under the some bandwidth. So, it is certified that the hypercube computer is suitable for the multicast communication servers supporting the various groups.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.2 no.1
• /
• pp.17-32
• /
• 1992

The multi-layer neural network that has broadly been utilized in designing the controller of robot manipulator possesses the desirable characteristics of learning capacity, by which the uncertain variation of the dynamic parameters of robot can be handled adaptively, and parallel distributed processing that makes it possible to control on real-time. However the error back propagation algorithm that has been utilized popularly in the learning of the multi-layer neural network has the problem of its slow convergence speed. In this paper, an approach to improve the convergence speed is proposed using the fuzzy logic that can effectively handle the uncertain and fuzzy informations by linguistic level. The effectiveness of the proposed algorithm is demonstrated by computer simulation of PUMA 560 robot manupulator.

• 한국정보통신설비학회:학술대회논문집
• /
• 2008.08a
• /
• pp.389-393
• /
• 2008

OFDM (Orthogonal Frequency Division Multiplexing) modulation using the orthogonal subcarriers reduces the delay spread by increasing robustness to multipath fading and can use overlapped bandwidth due to orthogonality on frequency domain. Thus data rate and spectral efficiency are increased. Because of these reason, OFDM is used for high speed data transmission for multimedia transmission as HSDPA, WiBro, WLAN. However OFDM also has drawbacks that have the high PAPR (Peak to Average Power Ratio). This high PAPR takes place because of parallel processing a number of data at once using a FFT processor. By high PAPR, amplifier doesn't act in dynamic range, so that BER performance is worse. In this paper, we reduce the PAPR using SLM(Selective Mapping). SLM doesn't effect on BER performance, but should transmit the side information for demodulation [2]. Also PAPR is higher as the number of FFT processor is larger. Thus SLM has high complexity. In this paper, we analyze the performance of SLM using scaling for no side information.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.2
• /
• pp.107-112
• /
• 2013

With the progress of NGS technologies, large genome data have been exploded recently. To analyze such data effectively, the assistance of HPC technique is necessary. In this paper, we organized a genome analysis pipeline to call SNP from NGS data. To organize the pipeline efficiently under HPC environment, we analyzed the CPU utilization pattern of each pipeline steps. We found that sequence alignment is computing centric and suitable for parallelization. We also analyzed the performance of parallel open source alignment tools and found that alignment method utilizing many-core processor can improve the performance of genome analysis pipeline.

• Proceedings of the KSR Conference
• /
• 2011.10a
• /
• pp.2885-2890
• /
• 2011

Simulation study were performed for fire location effect on the smoke spread in the deeply-underground subway station(DUSS). In this research, Shingumho station (The line # 5, Depth: 46m) has been selected as case-study for the analysis of smoke-spread effect with the different fire location. Field test data measured for actual fan in DUSS was applied as a condition of a simulation. The whole station was covered in this analysis and 4 million grids were generated for this simulation. The fire driven flow was analyzed case by case to compare the smoke-spread effect according to the fire location. In order to enhance the efficiency of calculation, parallel processing by MPI was employed and LES(large eddy simulation) method in FDS code was adopted.

• KIPS Transactions on Computer and Communication Systems
• /
• v.3 no.1
• /
• pp.1-6
• /
• 2014

Most of the previous algorithms focus on computing the convex hull for a set of points. In this paper, we present a method for approximating the convex hull for a set of spheres with various radii in discrete space. Computing the convex hull for a set of spheres is a base technology for many applications that study structural properties of molecules. We present a voxel map data structures, where the molecule is represented as a set of spheres, and corresponding algorithms. Based on CUDA programming for using the parallel architecture of GPU, our algorithm takes less than 40ms for computing the convex hull of 6,400 spheres in average.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2004.08a
• /
• pp.72-82
• /
• 2004

The deformation mechanism of hexagonal close-packed materials is quite complicate including slips and twins. A deformation mechanism, which accounts for both slip and twinning, was investigated for polycrystalline hop materials. The model was developed in a finite element polycrystal model formulated with initial strain method where the stiffness matrix in FEM is based on the elastic modulus. We predicted numerically the texture of Mg alloy(AZ31B) sheet by using FEM based on crystal plasticity theory. Also, we introduced the recrystallized texture employed the maximum energy release theory after rolling. From the numerical study, it was clarified that the shrink twin could not be the main mechanism for shortening of c-axis, because the lattice rotation due to twin rejects fur c-axis to become parallel to ND(normal direction of plate). It was showed that the deformation texture with the pyramidal slip gives the ring type pole figure having hole in the center.

• Journal of KIISE:Databases
• /
• v.27 no.1
• /
• pp.79-89
• /
• 2000

For processing signature file in parallel, an effective signature file declustering method is needed. The Linear Code Decomposition Method(LCDM) used for the Hamming Filter may give a good performance in some cases, but due to its static property, it fails to evenly decluster signature file when signature are skewed. In addition, it has other problems such as limited scalability and non-determinism. In this paper we propose a new signature file declustering method, called Inner-product method, which overcomes those problems in the LCDM. The Inner-product method declusters signature file dynamically based on the signature difference which is computed by using signature inner product. we show through the simulation experiment that the Inner-product outperforms the LCDM under various data workloads.