http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
오명훈,Oh, Myeong-Hoon 한국인터넷방송통신학회 2021 한국인터넷방송통신학회 논문지 Vol.21 No.6
마이크로서버는 컴퓨팅 서버의 일종으로 2개 이상의 CPU 소켓을 별도의 컴퓨팅 보드에 구현하고, 다수 개의 컴퓨팅 보드를 메인 보드에 집적하는 형태를 지닌다. 클러스터 시스템을 구축하는데 있어서, 마이크로서버를 사용하면 기존의 서버를 여러 대 랙에 장착하는 방법에 비해, 에너지 효율, 상면, 관리 용이성 측면에서 이점이 있다. 또한, 마이크로서버는 컴퓨팅 보드내 CPU 소켓들, 혹은 컴퓨팅 보드끼리 별도의 내부 연결망을 사용할 수 있어서 성능 측면에서도 이점이 존재한다. 본 논문에서 제안된 마이크로서버는 4개의 CPU 소켓을 지닌 컴퓨팅 보드를 메인보드에 총 16개 장착할 수 있는 서버로 Serial-RapidIO (SRIO)를 내부 연결망으로 사용한다. 마이크로서버의 핵심 성능 이슈인 내부 연결망 측면에서의 성능 비교 분석을 위해, 상용 마이크로서버와 내부 연결망 성능을 비교하고 정량화한다. 시험 결과, 내부 연결망을 활용한 데이터 전송 시 대역폭 측면에서 최대 7배 높은 성능을 보였다. 아울러, 실제 클라우드 컴퓨팅에 사용되는 벤치마크 프로그램 적용 시험에서도 유사 CPU 성능 마이크로서버 대비 60%의 수행시간 감소 효과를 얻었다. A microserver is a type of a computing server, in which two or more CPU nodes are implemented on a separate computing board, and a plurality of computing boards are integrated on a main board. In building a cluster system, the microserver has advantages in several points such as energy efficiency, area occupied, and ease of management compared to the existing method of mounting legacy servers in multiple racks. In addition, since the microserver uses a fast interconnection network between CPU nodes, performance improvement for data transfers is expected. The proposed microserver can mount a total of 16 computing boards with 4 CPU nodes on the main board, and uses Serial-RapidIO (SRIO) as an interconnection network. In order to analyze the performance of the proposed microserver in terms of the interconnection network which is a core performance issue of the microserver, we compare and quantify the performance of commercial microservers. As a result of the test, it showed up to about 7 times higher bandwidth improvement when transmitting data using the interconnection network. In addition, with CloudSuite benchmark programs used in actual cloud computing, maximum 60% reduction in execution time was obtained compared to commercial microservers with similar CPU performance specification.
비동기식 프로세서 설계 기법을 통한 저전압 동작 특성 분석과 전자기파 방사량 측정 분석
오명훈(Myeong-Hoon Oh),김학영(Hag-Young Kim),이정근(Jeong-Gun Lee) 한국정보기술학회 2014 한국정보기술학회논문지 Vol.12 No.2
In an asynchronous circuitry, the current consumption is distributed over time and, hence, its corresponding emission of electro-magnetic interference (EMI) can be reduced. Moreover, an asynchronous circuitry can guarantee the functionality at lower operating voltage than nominal supply voltage due to the absence of a global clock signal. To evaluate such characteristics of low-voltage operations and EMI emission in asynchronous circuits, we implemented synchronous and asynchronous MIPS processor cores on a commercial FPGA and measured EMI emission and performance of both versions. EMI emission peaks were observed at every harmonic frequency bands of 50MHz clock frequency and the maximum emission peak of the asynchronous MIPS (62.08dBμV) is lower than the synchronous one (67.25dBμV). It is also observed that the asynchronous MIPS is superior to the synchronous counterpart in terms of reliability in the experiment using lower operating voltage (3.2V) than nominal voltage (5.V).
그래픽 하드웨어 기반의 Full HD 급 고화질 클라우드 데스크톱 서비스 프로토타입 구현
오명훈(Myeong-Hoon Oh),박찬호(Chanho Park),권원옥(Wonok Kwon),김성운(Seongwoon Kim) 한국정보기술학회 2015 한국정보기술학회논문지 Vol.13 No.11
In cloud desktop services, a service provider provides a service customer with virtual desktop environment on a remote server through network. These services are gradually requiring high definition based services like high quality videos and 3D contents applications as well as basic operations such as an office environment and web browsing previously used in conventional PCs. This paper suggests a fundamental architecture and describes an implementation of a prototype for the cloud desktop service that supports the capabilities for full high-definition applications. The prototype is based on the graphic hardwares that enable remote video and USB information in a server side to be delivered to user devices in a client side through network. With the implemented prototype, a cloud desktop service was demonstrated in a LAN environment. It was also confirmed that the prototype exhibits a stable performance from a result of experimental measurement for response time.
마이크로서버용 고속 패브릭 연결망을 위한 SRIO 브리지 통신 구현
오명훈(Myeong-Hoon Oh),신지호(Ziho Shin),권혁제(Hyukje Kwon),김영우(Young Woo Kim),권원옥(Won-Ok Kwon),김학영(Hag-Young Kim) 대한전자공학회 2015 대한전자공학회 학술대회 Vol.2015 No.11
In the era of cloud computing, various server systems have been introduced. A microserver, which has low-power and high-density characteristics, is known to be a novel approach of future server systems. The microserver can include multiple computing resources and, hence, a high-speed and reliable fabric interconnection among the computing resources is highly required in order to utilize those resources. In this paper, for the implementation of the fabric interconnection mentioned above, Serial RapidIO is introduced. We also address the end-point network environment and suggest the test result for feasibility on Linux operating system.
신치훈(Chi-Hoon Shin),오명훈(Myeong-Hoon Oh),박경(Kyoung Park),김성운(Sung-Woon Kim) 대한전기학회 2006 정보 및 제어 심포지엄 논문집 Vol.2006 No.1
In this paper we present a new-generation sensor network processor which is not optimized in circuit level, but in system architecture level. The new design build on a conventional processor architecture, improving the design by focusing on application oriented specification, ISA, and micro-architectural optimization that reduce overall design size and advance energy-per-instruction. The design employs harvard architecture, 8-bit data paths, and an compact 19 bit wide RISC ISA. The design also features a unique interrupt handler which offloads periodical monitoring jobs from the main part of CPU. Our most efficient design is capable of running at 300 ㎑ (0.3 MIPS) while consuming only about few pJ/instruction.
권혁제(Hyukje Kwon),오명훈(Myeong-Hoon Oh),신지호(Ziho Shin),김학영(Hag-Young Kim),김영우(Young-Woo Kim) 대한전자공학회 2016 대한전자공학회 학술대회 Vol.2016 No.11
In this paper, we described modules that can be configured 256Gbs bandwidth at minimum as an interconnection on a micro-server. In this module, we used the SRIO on networking to get the merit which is a flexibility in configuring the networks. Micro-server was configured with sixteen compute cards, each card was composed with 4 CPU. The interconnection module connected to two compute cards. Each CPU supports PCIe gen1 2.5Gbps, 1x at minimum.