• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.5
• /
• pp.1342-1348
• /
• 1998

The major limitation of MOM solution has alaways been the computer CPU time and storage size, needed to carry out the impedance matrix computation. A new formulation technique using Sorce Superimposion method is presented in order to cut down computerstorage requirements and CPU time based on the equivalence principle and induction theorem. The numerical results are shown to give good agreement to those calculated by the conventional method and also application example is presented.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.32-35
• /
• 2017

직접 통로(Direct Pass-through) 기반 GPU(Graphic Processing Unit) 가상화 기법은 클라우드 환경에서 가상머신에게 GPU 장치의 기능을 지원하기 위한 일반적인 방법 중 하나이다. GPU 장치는 GPGPU 기술을 통해 연산을 가속화 할 수 있기 때문에 클라우드 환경에서도 가상머신에 고성능 연산을 지원하기 위해 많이 사용되고 있다. 하지만 기존 가상머신 스케줄링 기법은 가상머신의 CPU 사용 시간을 기반으로 스케줄링 되며, GPU 자원 사용을 고려하지 않는다. 본 논문에서는 GPU와 CPU 연산을 수행하는 가상머신들이 동시에 실행되는 환경에서 성능 실험을 통해 가상머신의 GPU 연산이 다른 가상머신에게 미치는 성능 영향과 GPU 작업 길이가 다른 가상머신에게 미치는 영향을 분석한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06d
• /
• pp.4-6
• /
• 2012

모바일 환경에서는 배터리 용량의 한계로 에너지 효율성이 매우 중요하나, CPU의 이용률에 기반한 DVFS로는 소비 전력 감소에 한계가 존재한다. 모바일 어플리케이션은 사용자와의 상호 동작에 따라 다른 성향을 보여주기 때문에, 이 특성을 이용하여 사용자가 체감하는 성능저하 없이 소비 전력을 줄일 수 있다. 본 논문에서는 사용자 입력과 어플리케이션 간의 상관 관계를 바탕으로 수행 단계를 분석하였으며, 이에 맞추어 CPU 동작 주파수 범위를 조절함으로써 사용자 반응성을 유지하면서 소비 전력을 줄이는 방법을 제안한다. 이를 안드로이드 플랫폼에 적용하여 실험한 결과, 기존 대비 휴대폰의 이용 시간을 최대 2시간 증가시켰다.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.5
• /
• pp.91-98
• /
• 2004

Human uses level difference and time difference to get space information. Therefore this paper shows that method to presume direction of sound source by time difference and to mark presumed position. The position means direction from geometrical center of sensors to the sound source. To get the time difference of microphones input level, we will be explained about arrangement of microphones which used for the sensor to take the sound signal. It is included distance among the 3 microphones and distance between microphones and sound source. Secondly, input signals are transmitted to CPU througth digital process. CPU is used to DSP(Digital Signal Processor) for manage the signal by real time. Finally, the position of sound source is perceived by an explained algorithm in this paper.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.2
• /
• pp.127-133
• /
• 2019

Since frequent item extraction so far requires searching for patterns and traversal for the FP-Tree, it is more likely to store the mining data in a tree and thus CPU time is required for its searching. In order to overcome these drawbacks, in this paper, we provide each item with its location identification of transaction data without relying on conditional FP-Tree and convert transaction data into 2-dimensional position information look-up table, resulting in the facilitation of time and spatial accessibility. We propose an algorithm that considers the mapping scheme between the location of items and items that guarantees the linear time complexity. Experimental results show that the proposed method can reduce many execution time and memory usage based on the data set obtained from the FIMI repository website.

• Proceedings of the KIEE Conference
• /
• 2002.11b
• /
• pp.218-220
• /
• 2002

본 논문은 시간영역 모의 테크닉과 과도안정도 해석 직접법을 결합한 하이브리드형 방법을 설명한다. 먼저 객체지향기법을 이용한 시간영역 모의를 통해 실제 시스템의 궤도를 계산하고, 과도안정도 여유도를 계산하기 위한 안정도 지수를 유도하기 위해 과도 에너지함수를 평가한다. 일단 시스템의 어떤 상태(안정 또는 불안정)가 확인되면 CPU 시간을 줄이기 위해 시간영역 모의를 중단하기 위한 적절한 기준이 제시되었다.

• Composites Research
• /
• v.15 no.6
• /
• pp.1-7
• /
• 2002

In VARTM process where a sacrificial medium is used to facilitate the resin flow, the velocity of resin varies drastically between the sacrificial medium and the fiber preform. Although the thickness-to-length ratio of a VARTM product is usually small, a 3-D analysis is prerequisite to analyze the lead-lag flow in the two different media. The problem associated with the full 3-D analysis is the CPU time. A full 3-D numerical mesh comprising large number of nodes requires an impractical CPU time on average computer platforms. In this study, a dual-scale analysis technique was developed. The flow analysis for the entire calculation domain was conducted in 2.5-D, and the 3-D analysis was performed for a small area of special concern. In some numerical examples, the local 3-D analysis could discover an eccentric flow pattern as well as the lead-lag flow that will inevitably be neglected in 2.5-D simulations. The global-local analysis technique practiced in this study can be used to analyze the intricate flow of resin through non-uniform media in affordable CPU times.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.21 no.5
• /
• pp.357-370
• /
• 2009

EFDC, ESCORT and MIKE21 models are applied at the Gomso Bay to investigate each models' facilities of tidal flat simulation. Comparisons with observation data show that all models simulate hydrodynamic phenomena and tidal flat well. CPU time and WCM are examined to evaluate the efficiency of the models, and the effects of flooding/drying depth and bottom friction are examined to analyze models' facilities of simulating tidal flat. The EFDC model is considered to be fairly good in accuracy, stability and applicability, it is, however, poor in efficiency and its complexity. While the ESCORT model is superior to the EFDC in simulation of tidal flat, it is inferior to the EFDC in CPU time and simulation of bottom friction. The MIKE21 model is excellent in efficiency, but some numerical noise would be detected at low water, not permitting correction of the model.

• Journal of KIISE:Software and Applications
• /
• v.32 no.2
• /
• pp.85-98
• /
• 2005

Recently, a lot of researches about mobile vision using Personal Digital Assistant(PDA) has been attempted. Many CPUs for PDA are integer CPUs, which have no floating-computation component. It results in slow computation of the algorithms peformed by vision system or image processing, which have much floating-computation. In this paper, in order to resolve this weakness, we propose the Client(PDA)/server(PC) architecture which is connected to each other with a wireless LAN, and we construct the system with pipelining processing using two CPUs of the Client(PDA) and the Server(PC) in image sequence. The Client(PDA) extracts tentative text regions using Edge Density(ED). The Server(PC) uses both the Multi-1.aver Perceptron(MLP)-based texture classifier and Connected Component(CC)-based filtering for a definite text extraction based on the Client(PDA)'s tentativel99-y extracted results. The proposed method leads to not only efficient text extraction by using both the MLP and the CC, but also fast running time using Client(PDA)/server(PC) architecture with the pipelining processing.

• The KIPS Transactions:PartA
• /
• v.10A no.4
• /
• pp.295-304
• /
• 2003

Active object is an Independent runnable unit which is scheduled by CPU in creation time. In this paper, we define the active object and suggest ART(Active object RunTime support systems) which controls creation and execution of the active object. ART can Provide users locational transparency and support easy method call mechanism. We also designed a communication model among active objects and implemented a communication method to make the distributed programing possible. The implementation target platform of ART is an embedded system which has only limited resources and runs in the distributed computing environment.