• 전자통신동향분석
• /
• 제39권1호
• /
• pp.62-73
• /
• 2024

Artificial intelligence relies on data-driven analysis, and the data processing performance strongly depends on factors such as memory capacity, bandwidth, and latency. Fast and large-capacity memory can be achieved by composing numerous high-performance memory units connected via high-performance interconnects, such as Compute Express Link (CXL). CXL is designed to enable efficient communication between central processing units, memory, accelerators, storage, and other computing resources. By adopting CXL, a composable computing architecture can be implemented, enabling flexible server resource configuration using a pool of computing resources. Thus, manufacturers are actively developing hardware and software solutions to support CXL. We present a survey of the latest software for CXL memory utilization and the most recent CXL memory emulation software. The former supports efficient use of CXL memory, and the latter offers a development environment that allows developers to optimize their software for the hardware architecture before commercial release of CXL memory devices. Furthermore, we review key technologies for improving the performance of both the CXL memory pool and CXL-based composable computing architecture along with various use cases.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1553-1557
• /
• 2005

High performance and reliability of the storage system to handle a very large amount of data has been become very important. Many techniques have been applied on the various application systems to establish very large capacity storage that satisfy the requirement of high I/O speed and physical or logical failure protection. We applied RAID and LVM to construct a storage system for the global data center which needs a very reliable large capacity storage system. The storage system is successfully established and equipped on the latest Linux application server.

• 경영과학
• /
• 제11권1호
• /
• pp.39-50
• /
• 1994

Carousel systems can be classified into unidirectional and bidirectional carousel, depending on the hardware structure. Bidirectional carousels again can be classified into reversible and irreversible carousel, depending on whether the reversal of direction is allowed or not during the process of an order. In this study, single carousel served by one operator is considered. Analytical models for the rotation distance to process an order of size are developed for unidirectional and irreversible bidirectional cases. For reversible bidirectional and Nearest Neighbor Heuristic cases, simulation results of the average rotation distance are presented. Since the throughput capacity of carousel systems also depends upon the picking time of the operator; order size; and the rotation speed of carousel, the model considers the above three factors to express the throughput capacity of single carousel systems.

• 반도체디스플레이기술학회지
• /
• 제10권4호
• /
• pp.25-29
• /
• 2011

Emerging demands for rechargeable battery for various applications needs more effective battery management system such as the prediction of the usable time about a battery. Many prediction methods have been suggested but none of them come into bounds of reliability. In this paper, we proposed a new prediction algorithm for the remaining capacity of a rechargeable battery by using the transformed curve based on its impedance. Hardware for monitoring a battery was designed and made. Through a series of experiment, we showed the effectiveness of the proposed prediction algorithm of a battery's remaining capacity.

• 전자공학회논문지
• /
• 제53권4호
• /
• pp.79-86
• /
• 2016

물리계층보안은 신호의 물리적 특성을 이용하여 정보를 보호하는 보안 기법이다. 현재 이에 대한 연구가 활발히 진행 중이지만 해결해야할 다음과 같은 문제점들이 존재한다. 도청자는 자신의 존재를 숨기기 위해서 자신의 채널상태정보를 다른 합법적인 노드와 공유하지는 않는다. 그리고 노드가 신호를 전송할 때 하드웨어 왜곡이 발생하지만 많은 연구들은 노드 모델들이 이상적인 것으로 가정을 하고, 하드웨어 왜곡문제를 고려하지 않고 있다. 이와 같은 문제점들을 해결하기 위한 본 논문의 주요한 특징 및 기여도는 다음과 같다. 첫째, 도청자의 채널상태정보를 얻기 위해서 조력자노드를 합법적인 노드주변에 설치하고, 조력자노드의 채널상태정보를 이용하여 노드모델에서 하드웨어 왜곡을 고려한다. 둘째, 제안된 시스템 모델의 인터셉트 확률에 대한 Closed-Form Expression을 제시한다. 제안된 시스템의 성능평가를 위해서 다양한 시뮬레이션를 통하여 제안된 시스템 모델의 물리계층보안에 미치는 영향을 알아본 결과, 불완전한 채널상태정보는 인터셉트 확률에는 영향을 미치지 못한 반면에, 불완전한 노드모델의 경우, 인터셉트 확률, 에르고딕 시크리스 용량과 보안채널용량에 영향을 준다는 것을 보여준다.

• 조명전기설비학회논문지
• /
• 제23권1호
• /
• pp.59-68
• /
• 2009

분산진원으로서 연료전지 발전장치는 100w부터 수백[kw]의 용량을 가지며 종전의 대규모 전력설비와 비교하여 높은 신뢰도를 갖는 고품질의 전력을 공급할 수 있다. 본 연구에서는 소형 분산전원으로서 PEMFC(polymer electrolyte membrane fuel cell)연료전지 발전장치에 대한 PSIM(power electronics simulation tool) 모델을 설정하고 이를 바탕으로 DSP(digital signal processor)기반의 연료전지 하드웨어 시뮬레이터를 구현하였다. 연료전지 전류와 출력전압과의 관계는 연료전지의 전압-전류(V-I) 곡선 중 ohmic영역에서 1차 함수로 간략화 하였다. 구현된 시스템은 PEMFC 하드웨어 시뮬레이터, 절연형 풀 브리지 직류 부스트 컨버터 그리고 60[Hz] PWM인버터로 구성되어있다. 부하변동 및 과도상태에 대한 연료전지 하드웨어 시뮬레이터의 전압-전류-전력(V-I-P) 특성을 파악하였으며, 저항 부하 및 비선형 부하에 대한 전력변환기의 60[Hz] 정현파 교류출력 전압파형을 고찰하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권8호
• /
• pp.2450-2463
• /
• 2024

Many legacy information systems are currently being clouded. This is due to the advantage of being able to respond flexibly to the changes in user needs and system environment while reducing the initial investment cost of IT infrastructure such as servers and storage. The infrastructure of the information system migrated to the cloud is being integrated through the API connections, while being subdivided by using MSA (Micro Service Architecture) internally. DBMS (Database Management System) is also becoming larger after cloud migration. Scale calculation in most layers of the application architecture can be measured and calculated from auto-scaling perspective, but the method of hardware scale calculation for DBMS has not been established as standardized methodology. If there is an error in hardware scale calculation of DBMS, problems such as poor performance of the information system or excessive auto-scaling may occur. In addition, evaluating hardware size is more crucial because it also affects the financial cost of the migration. CPU is the factor that has the greatest influence on hardware scale calculation of DBMS. Therefore, this paper aims to calculate the conversion factor for CPU scale calculation that will facilitate the cloud migration between heterogeneous DBMS. In order to do that, we utilize the concept and definition of hardware capacity planning and scale calculation in the on-premise information system. The methods to calculate the conversion factor using TPC-H tests are proposed and verified. In the future, further research and testing should be conducted on the size of the segmented CPU and more heterogeneous DBMS to demonstrate the effectiveness of the proposed test model.

• 센서학회지
• /
• 제12권4호
• /
• pp.164-169
• /
• 2003

최근에 남북철도나 대륙횡단열차 등에 대한 사업성 타당조사가 이뤄지면서 철도차량에 대한 관심이 높아지고 있다. 타교통수단과 비교하여 철도차량은 환경친화적, 정시성, 신뢰성, 그리고 안전성의 특성이 있다. 우리는 한국형 고속철도차량을 평가하기 위한 하드웨어와 소프트웨어를 제작하였다. 소프트웨어는 측정된 데이터를 분석할 수 있도록 해준다. 이 논문에서는 센서를 사용한 철도차량 전기장치 측정에 관해서 연구하였다. 이 시험을 하기 위해서 신호변환시스템을 개발하였다. 이 계측시스템을 이용해서 가선전압, battery전압, 차축속도 등에 관한 중요한 시험결과를 얻을 수 있었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.575-580
• /
• 2003

In this paper, we develop a control hardware such as an FPGA based general purpose controller with a DSP board to solve nonlinear control problems. PID control algorithms are implemented in an FPGA and neural network control algorithms are implemented in a DSP board. PID controllers implemented on an FPGA was designed by using VHDL to achieve high performance and flexibility. By using high capacity of an FPGA, the additional hardware such as an encoder counter and a PWM generator, can be implemented in a single FPGA device. As a result, the noise and power dissipation problems can be minimized and the cost effectiveness can be achieved. In order to show the performance of the developed controller, it was tested for controlling nonlinear systems such as an inverted pendulum.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2005년도 추계종합학술대회
• /
• pp.1003-1006
• /
• 2005

Vertex shader of GPU in personal computer is advanced in functions as to be half of traditional fixed T&L functions. And, capacity of memory for saving resources to process instructions is unlimited. GPU that can be programmed by programmer is needed for mobile system as well as personal computer. In this paper, we implement software virtual machine for vertex shader using C++ Language. Our goal is designing hardware GPU that can apply to mobile system. The virtual machine consists of nVidia GPU instructions. Input Data to virtual machine is generated by Microsoft fxc compiler. That is to say, Input Data is compiled shader program written in HLSL, Cg, or ASM. The virtual machine will be a reference model for designing hardware GPU and can be used for Testbed to test added or modified instruction.