http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
MPCore Based Task Scheduling under Peak Power Constraint
Sunghwan Park,Byunggyu Ahn,Junmo Jung,Hyunglae Roh,Bong-sik Sihn,Liao ZhiRui,Jongwha Chong 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7
In this paper, we propose a new task scheduling algorithm which can chedule tasks under peak power constraint for MPCore systems. To gain accurate power profile of each task, we simulate the tasks under virtual platform which can estimate power consumption of each instruction. And then, we apply PWL (Piece-Wise Linear) modeling to the power profile of each task to divide a task to some sub-tasks. To meet peak power and average power constraints, we schedule each sub-task in MPCore. If there is no dependency between tasks, we use DVS algorithm in each core to minimize power consumption. We demonstrate the proposed approaches using some benchmark applications.
강승민(SeungMin Kang),박병규(ByungGyu Park),이성우(SungWoo Lee),안익현(IkHyeon Ahn),조민규(MinGyu Jo),최명현(MyeongHeon Choe),강태원(TaeWon Kang) 한국정보기술학회 2020 Proceedings of KIIT Conference Vol.2020 No.10
코로나(Covid-19)로 의심되는 증상에 대한 서술을 수집하여 양성 서술과 음성 서술로 구분한다. 이렇게 구분한 데이터에 대해 각 서술에 단어가 출현하는 빈도를 계산하여 베이즈 분류기를 만든다. 이렇게 생성한 분류기로 새로운 증상에 대한 서술을 분류하여 환자의 감염 여부를 판단한다. 실험 결과 학습 데이터의 89%, 검증데이터의 52%를 분류할 수 있었다. Descriptions of symptoms suspected of coronavirus (Covid-19) are collected and classified into positive and negative statements. For the separated data, the Bayesian classifier is created by calculating the frequency of occurrence of words in each description. With the classifier created in this way, the description of the new symptom is classified to determine whether the patient is infected. As a result of the experiment, 89% of the learning data and 52% of the verification data could be classified.