• Title/Summary/Keyword: Explicit Code

Search Result 190, Processing Time 0.025 seconds

Parallel LDPC Decoding on a Heterogeneous Platform using OpenCL

  • Hong, Jung-Hyun;Park, Joo-Yul;Chung, Ki-Seok
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.10 no.6
    • /
    • pp.2648-2668
    • /
    • 2016
  • Modern mobile devices are equipped with various accelerated processing units to handle computationally intensive applications; therefore, Open Computing Language (OpenCL) has been proposed to fully take advantage of the computational power in heterogeneous systems. This article introduces a parallel software decoder of Low Density Parity Check (LDPC) codes on an embedded heterogeneous platform using an OpenCL framework. The LDPC code is one of the most popular and strongest error correcting codes for mobile communication systems. Each step of LDPC decoding has different parallelization characteristics. In the proposed LDPC decoder, steps suitable for task-level parallelization are executed on the multi-core central processing unit (CPU), and steps suitable for data-level parallelization are processed by the graphics processing unit (GPU). To improve the performance of OpenCL kernels for LDPC decoding operations, explicit thread scheduling, vectorization, and effective data transfer techniques are applied. The proposed LDPC decoder achieves high performance and high power efficiency by using heterogeneous multi-core processors on a unified computing framework.

A Parameter Study for Static and Dynamic Denting

  • Jung, Dong-Won;Worswick, M.J.
    • Journal of Mechanical Science and Technology
    • /
    • v.18 no.11
    • /
    • pp.2009-2020
    • /
    • 2004
  • A parametric study of the factors controlling static and dynamic denting, as well as local stiffness, has been made on simplified panels of different sizes, curvatures, thicknesses and strengths. Analyses have been performed using the finite element method to predict dent resistance and panel stiffness. A parametric approach is used with finite element models of simplified panels. Two sizes of panels with square plan dimensions and a wide range of curvatures are analysed for several combinations of material thickness and strength, all representative of auto-motive closure panels. Analysis was performed using the implicit finite element code, LS-NIKE, and the explicit dynamic code, LS-DYNA for the static and dynamic cases, respectively. Panel dent resistance and stiffness behaviour are shown to be complex phenomena and strongly interrelated. Factors favouring improved dent resistance include increased yield strength and panel thickness. Panel stiffness also increases with thickness and with higher curvatures but decreases with size and very low curvatures. Conditions for best dynamic and static dent performance are shown to be inherently in conflict ; that is, panels with low stiffness tend to perform well under impact loading but demonstrate inferior static dent performance. Stiffer panels are prone to larger dynamic dents due to higher contact forces but exhibit good static performance through increased resistance to oil canning.

Numerical Simulation of Thin Sheet Metal Forming Process using Electromagnetic Force (전자기력을 이용한 박판 성형공정의 해석적 연구)

  • Seo, Y.H.;Heo, S.C.;Ku, T.W.;Song, W.J.;Kang, B.S.;Kim, J.
    • Transactions of Materials Processing
    • /
    • v.17 no.1
    • /
    • pp.35-45
    • /
    • 2008
  • Electromagnetic Forming (EMF) technology such as magnetic pulse forming, which is one of the high velocity forming methods, has been used for the joining and forming process in various industry fields. This method could be derived a series of deformation of sheet metal by using a strong magnetic field. In this study, numerical approach by finite element simulation of the electromagnetic forming process was presented. A transient electromagnetic finite element code was used to obtain the numerical model of the time-varying currents that are discharged through the coil in order to obtain the transient magnetic forces. Also, the body forces generated in electromagnetic field were used as the loading condition to analyze deformation of thin sheet metal workpiece using explicit dynamic finite element code. In this study, after finite element analysis for thin sheet metal forming process with free surface configuration was performed, analytical approach for a dimpled shape by using EMF was carried out. Furthermore, the simulated results of the dimpled shape by EMF were compared with that by a conventional solid tool in view of the deformed shape. From the results of finite element analysis, it is confirmed that the EMF process could be applied to thin sheet metal forming.

Numerical Modeling for Cumulative Impact of Automotive Bumper (자동차 범퍼의 누적 충격 평가)

  • Kim, Heon-Young;Choi, Jong-Gil;Kim, Jung-Min;Lee, Kang-Wook;Yeo, Tae-Jung
    • Transactions of the Korean Society of Automotive Engineers
    • /
    • v.15 no.3
    • /
    • pp.29-34
    • /
    • 2007
  • Numerical analyses are carried out to evaluate the cumulative impact damage of an automotive front end bumper under the low speed crash events(CMVSS215) by using explicit code. Results of first impact simulation, which are deformed shape, thickness, stress tensors and strain tensors, are used as the initial conditions for a next impact simulation. Between the events, the residual vibration is damped out by using nodal damping, and then recovery after each event is evaluated by several methods, one of which is a springback analysis with implicite finite element analysis code. The coupled analysis scheme for the evaluation of cumulative impact damage is verified through the comparison with test results.

Parallel Stratified and Rotating Turbulence Simulation based on MPI (MPI 기반의 병렬 성층${\cdot}$회전 난류 시뮬레이션)

  • Kim, Byung-Uck;Yang, Sung-Bong
    • The Transactions of the Korea Information Processing Society
    • /
    • v.7 no.1
    • /
    • pp.57-64
    • /
    • 2000
  • We describe a parallel implementation for the large-eddy simulation(LES) of stratified and rotating turbulence based on MPI. The parallelization strategy is specified by eliminating the tridiagonal solver with explicit method and by domain decompositions for solving the poisson equation. In this simulation we have run on CRAY-T3E under the message passing platform MPI with a various domain decomposition and the scalability of this parallel code of LES are also presented. The result shows that we can gain up to 16 times faster speed up on 64 processors with xyz-directional domain decomposition and scalable up to $128{\times}128{\times}$ which processing time is almost similar to that of $40{\times}40{\times}40$ on a single processor machine with a sequential code.

  • PDF

Calculation of fuel temperature profile for heavy water moderated natural uranium oxide fuel using two gas mixture conductance model for noble gas Helium and Xenon

  • Jha, Alok;Gupta, Anurag;Das, Rajarshi;Paraswar, Shantanu D.
    • Nuclear Engineering and Technology
    • /
    • v.52 no.12
    • /
    • pp.2760-2770
    • /
    • 2020
  • A model for calculation of fuel temperature profile using binary gas mixture of Helium and Xenon for gap gas conductance is proposed here. In this model, the temperature profile of a fuel pencil from fuel centreline to fuel surface has been calculated by taking into account the dilution of Helium gas filled during fuel manufacturing due to accumulation of fission gas Xenon. In this model an explicit calculation of gap gas conductance of binary gas mixture of Helium and Xenon has been carried out. A computer code Fuel Characteristics Calculator (FCCAL) is developed for the model. The phenomena modelled by FCCAL takes into account heat conduction through the fuel pellet, heat transfer from pellet surface to the cladding through the gap gas and heat transfer from cladding to coolant. The binary noble gas mixture model used in FCCAL is an improvement over the parametric model of Lassmann and Pazdera. The results obtained from the code FCCAL is used for fuel temperature calculation in 3-D neutron diffusion solver for the coolant outlet temperature of the core at steady operation at full power. It is found that there is an improvement in calculation time without compromising accuracy with FCCAL.

Platform development for multi-physics coupling and uncertainty analysis based on a unified framework

  • Guan-Hua Qian;Ren Li;Tao Yang;Xu Wang;Peng-Cheng Zhao;Ya-Nan Zhao;Tao Yu
    • Nuclear Engineering and Technology
    • /
    • v.55 no.5
    • /
    • pp.1791-1801
    • /
    • 2023
  • The multi-physics coupled methodologies that have been widely used to analyze the complex process occurring in nuclear reactors have also been used to the R&D of numerical reactors. The advancement in the field of computer technology has helped in the development of these methodologies. Herein, we report the integration of ADPRES code and RELAP5 code into the SALOME-ICoCo framework to form a multi-physics coupling platform. The platform exploits the supervisor architecture, serial mode, mesh one-to-one correspondence and explicit coupling methods during analysis, and the uncertainty analysis tool URANIE was used. The correctness of the platform was verified through the NEACRP-L-335 benchmark. The results obtained were in accordance with the reference values. The platform could be used to accurately determine the power peak. In addition, design margins could be gained post uncertainty analysis. The initial power, inlet coolant temperature and the mass flow of assembly property significantly influence reactor safety during the rod ejections accident (REA).

Development of the LMFBR Accident Analysis Computer Code (고속증식로 사고분석 코드의 개발)

  • Hee Cheon No
    • Nuclear Engineering and Technology
    • /
    • v.16 no.2
    • /
    • pp.47-57
    • /
    • 1984
  • Mathematically-rigorous time-volume averaged conservation equations were simplified to established the differential equations of THERMIT-6S, which is a two-fluid 3-D code. The difference equations of THERMIT-6S were obtained by discretizing the proceeding set of differential equations. The spatial discretization is characterized by a first-order spatial scheme, donor cell method, and staggered mesh layout. For time discretization, a first order semi-implicit scheme treats implictly sonic terms and terms relating to local transport phenomena and explicitly convective terms. The results were linearized by the Newton-Raphson method. In order to construct the reduced pressure equation, the linearized equations were manipulated so that all variables are coupled between mesh cells through only the pressure variable. By simulating numerically the OPERA-15 experiment, it was found that THERMIT-6S is a very powerful code in predicting reactor behavior after sodium boiling including flow coastdown, reversal flow and flow oscillation.

  • PDF

Stability Analysis of Pile/Slope Systems Considering Pile-slope Interaction (억지말뚝-사면의 상호작용을 고려한 사면안전율 분석)

  • 김병철;유광호;정상섬
    • Proceedings of the Korean Geotechical Society Conference
    • /
    • 2003.03a
    • /
    • pp.513-520
    • /
    • 2003
  • A numerical comparison or predictions by limit equilibrium analysis and 3n analysis is presented for slope/pile system. Special attention is given to the coupled analysis based on the explicit-finite-difference code, FLAC. To this end, an internal routine (FISH) was developed to calculate a factor of safety for a pile-reinforced slope according to shear strength reduction technique. The case of coupled analyses was performed for stabilizing piles in slope in which the pile response and slope stability are considered simultaneously and subsequently the factors of safety are compared to uncoupled analysis (limit equilibrium analysis) solution for a homogeneous slope. Based on a limited parametric study, it is shown that in the free-head condition the factor of safety in slope is more conservative for a coupled analysis than for an uncoupled analysis and a definitely larger value represents when piles are installed in the middle of the slopes and are restrained in the pile head.

  • PDF

The Analysis of Tunnel Excavation using Finite Difference Method (유한 차분법을 이용한 터널 단면 해석)

  • 김동후;유건선
    • Proceedings of the Korean Geotechical Society Conference
    • /
    • 1991.10a
    • /
    • pp.166-178
    • /
    • 1991
  • Using FLAC, which is an explicit finite difference code written for analysis of problems in geotechnical engineering, a particular example 2 in the Korean Geothechnical Society News has been analysed. The elastoplasticity formulation in FLAC assumes an elastic, perfectly plastic solid in plane strain which conforms to a Mohr-Coulomb yield condition. During tunnel excavations by stages, stresses and diaplacements in region around the tunnel varies according to distance from the face of tunnel and installation of tunnel supports, and soon. In this analysis, the three dimensional support effect of the rock mass during the process of excavation is simulated by using the stress distribution method, and varying the material constants of shotcrete in each stage also considered. The maximum convergency is occured at the crown of the tunnel and estimated to be about 12mm.

  • PDF