고속 중형 컴퓨터 (주전산기 Ⅲ)의 검증 방법

윤석한(S H Yoon),한우종(W J Han),박장석(J. S. Park),박종원(C. W. Park),김창석(C S Kim),최효진(H J Choi) 한국정보과학회 1992 한국정보과학회 학술발표논문집 Vol.19 No.2

본 논문은 한국전자통신연구소에서 수행중인 고속 중형 컴퓨터 (주선산기Ⅲ) 개발 사업에 적용되고 있는 검증 방법(verification and validation)에 관한 것이다. 복잡하고 다양한 기능을 갖는 대규모 시스템 개발 사업의 효율적인 품질 향상을 위해서는 검증 활동이 필요하다. 이 글에서는 고속 중형 컴퓨터 개발에 적용되고 있는 검증 계획과 검증 수행 방법에 대하여 서술하고, 검증 결과에 대하여 실제 적용 사례와 경험을 기술하였다.

단어 음성 데이타의 수집 및 DB 구성 시스템

이용주(Y. J. LEE),한우종(W. J. HAN),김민택(M. T. KIM),류준형(J. H. RYU),박찬경(C. K. PARE),이의택(E. T. LEE),김경태(K. T. KIM),조철우(C. W. CHO) 대한전자공학회 1986 대한전자공학회 학술대회 Vol.1986 No.12

MPI 집합통신 성능 향상 연구 동향

안후영,박유미,김선영,한우종,H.Y., Ahn,Y.M., Park,S.Y., Kim,W.J., Han 한국전자통신연구원 2022 전자통신동향분석 Vol.37 No.6

Message Passing Interface (MPI) collective communication has been applied to various science and engineering area such as physics, chemistry, biology, and astronomy. The parallel computing performance of the data-intensive workload in the above research fields depends on the collective communication performance. To overcome this limitation, MPI collective communication technology has been extensively researched over the last several decades to improve communication performance. In this paper, we provide a comprehensive survey of the state-of-the-art research performed on the MPI collective communication and examine the trends of recently developed technologies. We also discuss future research directions for providing high performance and scalability to large-scale MPI applications.

CCIX 연결망과 메모리 확장기술 동향

김선영,안후영,전성익,박유미,한우종,Kim, S.Y.,Ahn, H.Y.,Jun, S.I.,Park, Y.M.,Han, W.J. 한국전자통신연구원 2022 전자통신동향분석 Vol.37 No.1

With the advent of the big data era, the memory capacity required for computing systems is rapidly increasing, especially in High Performance Computing systems. However, the number of DRAMs that can be used in a computing node is limited by the structural limitations of the hardware (for example, CPU specifications). Memory expansion technology has attracted attention as a means of overcoming this limitation. This technology expands the memory capacity by leveraging the external memory connected to the host system through hardware interface such as PCIe and CCIX. In this paper, we present an overview and describe the development trends of the memory expansion technology. We also provide detailed descriptions and use cases of the CCIX that provides higher bandwidth and lower latency than cases of the PCIe.

CXL 메모리 및 활용 소프트웨어 기술 동향

안후영,김선영,박유미,한우종,H.Y. Ahn,S.Y. Kim,Y.M. Park,W.J. Han 한국전자통신연구원 2024 전자통신동향분석 Vol.39 No.1

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.

SC20를 통해 본 HPC 기술 동향

어익수,모희숙,박유미,한우종,Eo, I.S.,Mo, H.S.,Park, Y.M.,Han, W.J. 한국전자통신연구원 2021 전자통신동향분석 Vol.36 No.3

High-performance computing (HPC) is the underpinning for many of today's most exciting new research areas, to name a few, from big science to new ways of fighting the disease, to artificial intelligence (AI), to big data analytics, to quantum computing. This report captures the summary of a 9-day program of presentations, keynotes, and workshops at the SC20 conference, one of the most prominent events on sharing ideas and results in HPC technology R&D. Because of the exceptional situation caused by COVID-19, the conference was held entirely online from 11/9 to 11/19 2020, and interestingly caught more attention on using HPC to make a breakthrough in the area of vaccine and cure for COVID-19. The program brought together 103 papers from 21 countries, along with 163 presentations in 24 workshop sessions. The event has covered several key areas in HPC technology, including new memory hierarchy and interconnects for different accelerators, evaluation of parallel programming models, as well as simulation and modeling in traditional science applications. Notably, there was increasing interest in AI and Big Data analytics as well. With this summary of the recent HPC trend readers may find useful information to guide the R&D directions for challenging new technologies and applications in the area of HPC.

CXL 인터커넥트 기술 연구개발 동향

김선영,안후영,박유미,한우종,S.Y. Kim,H.Y. Ahn,Y.M. Park,W.J. Han 한국전자통신연구원 2023 전자통신동향분석 Vol.38 No.5

With the widespread demand from data-intensive tasks such as machine learning and large-scale databases, the amount of data processed in modern computing systems is increasing exponentially. Such data-intensive tasks require large amounts of memory to rapidly process and analyze massive data. However, existing computing system architectures face challenges when building large-scale memory owing to various structural issues such as CPU specifications. Moreover, large-scale memory may cause problems including memory overprovisioning. The Compute Express Link (CXL) allows computing nodes to use large amounts of memory while mitigating related problems. Hence, CXL is attracting great attention in industry and academia. We describe the overarching concepts underlying CXL and explore recent research trends in this technology.

일별자료를 활용한 KOSPI 와 KOSDAQ 간의정보전달에 관한 실증 연구

김종출,황수석,김규호,신삼인 경남대학교 산업경영연구소 2019 지역산업연구 Vol.42 No.3

This paper studied the inter-relationship and market efficiency between KOSPI and KOSDAQmarket. The whole sample period is covered from January 4, 2000 to December 28, 2017. Weused the daily closing price data and trading volume of KOSDAQ and KOSPI. For these purposewe introduced Granger causality test based vector auto regressive analysis and the major empiricalresults are as follows;First, according to the unit root test, we find that the level variable and the log differencevariables are stationary with a statistically significant level. Second, we also find that there is a co-integration relationship between the level variables ofKOSPI and the KOSDAQ markets. Third, based on the Granger causality test, there is a feedback information transmissionbetween the returns of KOSDAQ and KOSPI but the impact of KOSPI is more dominant. Fourth, according to the impulse response analysis based on the vector auto regressive model,there is a similar relationship between the KOSPI and KOSDAQ returns. From these empirical results we infer that KOSPI is more efficient market than KOSDAQ andwe hope these results to helpful for the two market participants to make a investment decision andrisk management etc. 동 연구는 우리나라 KOSPI와 KOSDAQ에서의 일별 수익률과 거래 변화량 정보를 이용하여두 시장 간의 정보효율성에 대한 상호 인과관계를 여부를 분석하였다. 연구 기간으로는 2000년1월 4일부터 2017년 12월 28일까지이며, 총 4,446개의 KOSPI 및 KOSDAQ 일별 종가 자료와거래량 자료를 사용하였다. 연구 방법으로는 VAR모형에 기초한 그랜저(Granger) 인과관계 분석을 이용하였으며 주요 분석 결과는 다음과 같다. 첫째, KOSPI와 KOSDAQ의 수익률과 거래 변화량에 대한 분석을 위한 수준 변수와 차분 변수는 모두 안정적인 자료였으나, 상대적으로 차분 변수의 데이터가 보다 더 안정적인 자료로 나타나 차분변수를 활용하여 분석하였다. 둘째, KOSPI와 KOSDAQ의 각 변수 간에는 귀무가설이 통계적 유의한 수준에서 기각되었으며, 두 시장의 각 변수 간에는 장기적인 균형 관계가 존재함이 확인되었다. 셋째, VAR모형에 기초한 그랜저(Granger) 인과관계 분석결과, 두 시장의 수익률 간에는 상호영향을 미치는 피드백적인 영향이 있는 것으로 나타났으며, 그 영향의 크기는 KOSDAQ 수익률이 KOSPI 수익률에 미치는 영향력이 더 큰 것으로 나타났다. 넷째, 충격 반응에 대한 분석 결과에서는 KOSPI 수익률은 KOSDAQ 수익률에는 큰 영향을 받지 않는 것으로 나타났으나, KOSDAQ 수익률은 KOSPI 수익률과 자기 자신 모두에게 영향을 받는 것을 확인할 수 있었다. 이러한 분석 결과로 KOSPI와 KOSDAQ의 변동성은 KOSPI와 KOSDQ 두 시장 간에는 상호 영향을 미치는 피드백 적 영향을 미치고 있으며, 그 영향의 정도는 KOSPI가 KOSDAQ에 미치는 영향력이 상대적으로 더 큰 것을 확인할 수 있었다.