• ETRI Journal
• /
• v.45 no.6
• /
• pp.1035-1045
• /
• 2023

We propose a novel graphics processing unit (GPU) algorithm that can handle a large-scale 3D fast Fourier transform (i.e., 3D-FFT) problem whose data size is larger than the GPU's memory. A 1D FFT-based 3D-FFT computational approach is used to solve the limited device memory issue. Moreover, to reduce the communication overhead between the CPU and GPU, we propose a 3D data-transposition method that converts the target 1D vector into a contiguous memory layout and improves data transfer efficiency. The transposed data are communicated between the host and device memories efficiently through the pinned buffer and multiple streams. We apply our method to various large-scale benchmarks and compare its performance with the state-of-the-art multicore CPU FFT library (i.e., fastest Fourier transform in the West [FFTW]) and a prior GPU-based 3D-FFT algorithm. Our method achieves a higher performance (up to 2.89 times) than FFTW; it yields more performance gaps as the data size increases. The performance of the prior GPU algorithm decreases considerably in massive-scale problems, whereas our method's performance is stable.

• The KSFM Journal of Fluid Machinery
• /
• v.17 no.4
• /
• pp.47-52
• /
• 2014

An inducer is forward-attached to an impeller to improve the suction performance. This paper described the experimental and numerical investigations on the concept of NPSH similarity about the inducer scale. As Reynolds number decreased for the same scale inducer, the hydraulic performance is slightly reduced because of the viscosity. The suction performance similarity is in good agreement. For different scale inducers, the NPSH similarity did not follow the conventional rule which is proportional to the square of the inducer diameter. A cavity of two times scale inducer grows faster under cavitation inception, and the head is more drop as the fluid passes blades. Because of the simplified cavitation model and vapor pressure, the NPSH similarity dose not have an accuracy. This study suggested an empirical formula for the NPSH similarity.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.29 no.1
• /
• pp.47-56
• /
• 2015

Energy consumption in buildings is currently a real problem. That is why both assessment of energy performance and effective energy management including renewable energy system are essential. Thus, this paper focuses on a case study to analyze the energy performance and cooling & heating energy saving of a large scale shopping store in Daejeon city. The reference building is simulated by using TRNSYS dynamic simulation tool to examine its annual energy consumption. For annual energy analysis of building, one year energy consumption is surveyed in the field. The related study is carried out in large scale shopping store to investigate the energy consumption and energy use trend of heating, cooling, hot water, lighting, ventilation, equipments and other. The evaluation of energy performance of the geothermal heat pump system installed in a large scale shopping store is also analyzed by TRNSYS tool. From simulation results, it evaluated that the geothermal heat pump system is effective energy savings method in large scale shopping store.

• Journal of the Society of Naval Architects of Korea
• /
• v.52 no.6
• /
• pp.428-434
• /
• 2015

The objective of the present study is to analyze the results of the trial-test for the full-scale driving pump, which is arranged in the LCT (Large Cavitation Tunnel). Firstly, the reasons of selecting the final design pump are introduced in terms of the performance analysis in model tests. The trial-test items for the full-scale driving pump are measurements of output current/voltage at the inverter of the main motor and the flow velocity in the LCT test section. The test results show the increase in flow rate of about 10.7% and the decrease in pump head of about 26%, compared with those of final design-pump specification. The motor power has the margin of about 22%. The performance analysis for the full-scale pump is conducted using the commercial code (CFX-10). The delivered power calculated with CFX-10 shows good agreement with that extracted from the full-scale pump test. It is found that CFX-10 is useful to analyze a full-scale pump.

• Journal of the Optical Society of Korea
• /
• v.14 no.4
• /
• pp.388-394
• /
• 2010

In this paper the object recognition performance of a photon counting integral imaging system is quantitatively compared with that of a conventional gray scale imaging system. For 3D imaging of objects with a small number of photons, the elemental image set of a 3D scene is obtained using the integral imaging set up. We assume that the elemental image detection follows a Poisson distribution. Computational geometrical ray back propagation algorithm and parametric maximum likelihood estimator are applied to the photon counting elemental image set in order to reconstruct the original 3D scene. To evaluate the photon counting object recognition performance, the normalized correlation peaks between the reconstructed 3D scenes are calculated for the varied and fixed total number of photons in the reconstructed sectional image changing the total number of image channels in the integral imaging system. It is quantitatively illustrated that the recognition performance of the photon counting integral imaging system can be similar to that of a conventional gray scale imaging system as the number of image viewing channels in the photon counting integral imaging (PCII) system is increased up to the threshold point. Also, we present experiments to find the threshold point on the total number of image channels in the PCII system which can guarantee a comparable recognition performance with a gray scale imaging system. To the best of our knowledge, this is the first report on comparisons of object recognition performance with 3D photon counting & gray scale images.

• Proceedings of the KSME Conference
• /
• 2008.11b
• /
• pp.2548-2553
• /
• 2008

Prediction of the rotor blade performance is important for determining design factors such as weight and size in development of a small-scale helicopter. Generally, prediction of helicopter performance means the estimation of the power required for a given flight condition. However, due to lack of test data and analyzed results for small-scale rotor blade operated at low Reynolds numbers ($Re{\approx}10^5$), this is not an easy task. As an initial research, this work performs a modeling of a single rotor configuration with FLIGHTLAB and a experimental research with rotor test bed. In this process, we performed small-scale isolated single rotor by experimental and numerical method and achieved good agreement of the hover performance on the test data and simulation results.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.6 no.2
• /
• pp.380-391
• /
• 2014

In this study, the scale effect on the performance of the podded propeller of tractor type is investigated. Turbulent flow computations are carried out for Reynolds numbers increasing progressively from model scale to full scale using the CFD analysis. The result of the flow calculation for model scale Reynolds numbers agrees well with that of the experiment of a large cavitation tunnel. The existing numerical analysis indicates that the performance of the podded propeller blades is mainly influenced by the advance coefficient and relatively little by the Reynolds number. However, the drag of pod housing with propeller in operation is different from that of pod housing without propeller due to the acceleration and swirl of propeller slipstream which is altered by propeller loading as well as the pressure recovery and friction according to Reynolds number, which suggests that the pod housing drag under the condition of propeller in operation is the key factor of the scale effect on the performance between model and full scale podded propellers. The so called 'drag ratio', which is the ratio of pod housing drag to total thrust of podded propeller, increases as the advance coefficient increases due to accelerated flow in the slipstream of the podded propeller. However, the increasing rate of the drag ratio reduces continuously as the Reynolds number increases from model to full scale progressively. The contribution of hydrodynamic forces, which acts on the parts composed of the pod housing with propeller operating in various loading conditions, to the thrust and the torque of the total propeller unit are presented for a range of Reynolds numbers from model to full scales.

• Structural Engineering and Mechanics
• /
• v.90 no.5
• /
• pp.447-458
• /
• 2024

This study investigated the influence of scale ratio and vertical load on the seismic performance of Puzuo joints in traditional Chinese timber structures. Three low-cyclic reversed loading tests were conducted on three scaled specimens of Bujian Puzuo in Yingxian Wooden Pagoda. This study focused on the deformation patterns and analyzed seismic performance under varying scale ratios and vertical loads. The results indicated that the slip and rotational deformations of Bujian Puzuo were the primary deformations. The scale of the specimen did not affect the layer where the maximum interlayer slip occurred, but it did decrease the proportion of slip deformation. Conversely, the reducing vertical load caused the layer with the maximum slippage and the position of the damaged Dou components to shift upward, and the proportion of slip deformation increased. When the vertical load was decreased by 3.7 times, the maximum horizontal bearing capacity under positive and negative loadings, initial stiffness, and energy dissipation of the specimen decreased by approximately 60%, 58.79%, 69.62%, and 57.93%, respectively. The horizontal bearing capacity under positive loading and energy dissipation of the specimen increased by 35.63% and 131.54%, when the specimen scale was doubled and the vertical load was increased by 15 times.

• Journal of the Korean Solar Energy Society
• /
• v.22 no.2
• /
• pp.61-69
• /
• 2002

Daylighting is a powerful vehicle of architectural expression and provides buildings a living quality due to constantly changing properties in intensity and color. As an environmental system, daylighting should be subjected to the same level of rigorous analysis and review that any environmental system receives. For example, increasing cooling loads and causing visual displeasure are easily occurred owing to the excessive sunlight through windows. Therefore, daylighting performance attributes and physical characteristics must be described qualitatively and quantitatively in early design process. In various architectural daylighting performance evaluation methods, the scale model experiment can be applied due to it's simplicity and usefulness. So, this study aims to evaluate the validation of scale model experiments in architectural daylighting performance. For the purpose, two scale model experiments under clear sky are conducted. And, the validation of the experiments are evaluated by computer simulations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.6
• /
• pp.2824-2837
• /
• 2019

Container technologies are widely used in infrastructures to deploy and manage applications in cloud computing environment. As containers are light-weight software, the cluster of cloud applications can easily scale up or down to provide Internet-based services. Container-based applications can well deal with fluctuate workloads by dynamically adjusting physical resources. Current works of scheduling applications often construct applications' performance models with collected historical training data, but these works with static models cannot self-adjust physical resources to meet the dynamic requirements of cloud computing. Thus, we propose a self-adaptive automatic container scheduling framework AutoScale for cloud applications, which uses a feedback-based approach to adjust physical resources by extending, contracting and migrating containers. First, a queue-based performance model for cloud applications is proposed to correlate performance and workloads. Second, a fuzzy Kalman filter is used to adjust the performance model's parameters to accurately predict applications' response time. Third, extension, contraction and migration strategies based on predicted response time are designed to schedule containers at runtime. Furthermore, we have implemented a framework AutoScale with container scheduling strategies. By comparing with current approaches in an experiment environment deployed with typical applications, we observe that AutoScale has advantages in predicting response time, and scheduling containers to guarantee that response time keeps stable in fluctuant workloads.