http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
데이터 센터 전력 관리를 위한 부하 기반 전력 분석 모델 및 저비용 전력 예측 방법
최장호,강동오,이형직,배창석 한국차세대컴퓨팅학회 2014 한국차세대컴퓨팅학회 논문지 Vol.10 No.2
대부분의 데이터 센터는 평균적으로 총 자원의 20% ~ 30%만이 사용되고 있고 50% 미만의 전력 효율을 갖고 있 다. 유연한 전력 관리 정책을 적용하면 데이터 센터의 전력 효율을 높일 수 있고 이를 위해서는 동적 부하에 따른 정확한 전력 소비량을 예측 가능해야 한다. 하지만 부하에 따른 서버의 전력 소비량 예측은 시스템 구축비용 및 컴 퓨팅 비용이 매우 높은 편이다. 본 논문에서는 기존의 시스템 정보를 이용한 서버 전력 측정 방법을 소개하고 간편 한 구축과 저렴한 컴퓨팅 비용으로 시스템 부하에 따른 전력 소비량을 예측할 수 있는 방법을 제시한다. 시스템 부 하를 분류하고 부하 별 전력 모델을 도출하여 적용함으로써 보다 정확한 예측 방법을 제안하였다. 또한, 실제 서버 환경을 구축하여 평가를 진행하였고 제안한 방법의 타당성을 증명하였다.
최장호,곽찬희,이희석 한국경영정보학회 2017 Information systems review Vol.19 No.4
Analyzing and finding the risk factors in information technology (IT) projects have been discussed because risk management is an important issue in IT project management. This study obtained the risk factor checklists with priorities, analyzed the causal relationship of risk factors, and determined their influences on IT project management. However, only few studies systematically classified IT project risk factors in terms of risk exposure. These studies considered both the probability of occurrence and the degree of risk simultaneously. The present study determined 53 IT project risk factors on the basis of literature and expert group discussions. Additionally, this study presented clustering analysis based on the data of 140 project managers. The IT project risk factor classification framework was divided into four areas (HIHF, HILF, LIHF, and LILF). The present results can be used to help IT project managers establish effective risk management strategies and reduce IT project failures. This study also provides academic implication because it considers both the probability of occurrence and the degree of influence of risk factors.
DART: Fast and Efficient Distributed Stream Processing Framework for Internet of Things
최장호,박준용,박흰돌,민옥지 한국전자통신연구원 2017 ETRI Journal Vol.39 No.2
With the advent of the Internet-of-Things paradigm, the amount of data production has grown exponentially and the user demand for responsive consumption of data has increased significantly. Herein, we present DART, a fast and lightweight stream processing framework for the IoT environment. Because the DART framework targets a geospatially distributed environment of heterogeneous devices, the framework provides (1) an end-user tool for device registration and application authoring, (2) automatic worker node monitoring and task allocations, and (3) runtime management of user applications with fault tolerance. To maximize performance, the DART framework adopts an actor model in which applications are segmented into microtasks and assigned to an actor following a single responsibility. To prove the feasibility of the proposed framework, we implemented the DART system. We also conducted experiments to show that the system can significantly reduce computing burdens and alleviate network load by utilizing the idle resources of intermediate edge devices.