RISS 학술연구정보서비스

검색
다국어 입력

http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.

변환된 중국어를 복사하여 사용하시면 됩니다.

예시)
  • 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
  • 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
닫기
    인기검색어 순위 펼치기

    RISS 인기검색어

      검색결과 좁혀 보기

      선택해제
      • 좁혀본 항목 보기순서

        • 원문유무
        • 원문제공처
          펼치기
        • 등재정보
          펼치기
        • 학술지명
          펼치기
        • 주제분류
        • 발행연도
          펼치기
        • 작성언어

      오늘 본 자료

      • 오늘 본 자료가 없습니다.
      더보기
      • 무료
      • 기관 내 무료
      • 유료
      • 분산 시스템을 위한 트랜잭션 프로토콜 연구

        丘龍完 水原大學校 1991 論文集 Vol.9 No.-

        In this paper the transaction protocol which is able to perform efficiently communication between distributed process in computers linked to Local Area Network in environment of UNIX was proposed. This special protocol designed in kernel tasks, ie, transaction manager, transaction interrupt manager, and port manager.

      • TDX-집단전화 원격가입자 시스템의 구현

        구용완,황만수,권진혁 수원대학교 기초과학연구소 1992 基礎科學論文集 Vol.1 No.-

        This study has a purpose to design to supply the limited function of PABX widely for users. That is, in normal state this system give a service of customer premised system(CPS) and but in emergency supplies individual switching function as a stand-alone so that the service is available all the time. Also this system can locate in local area in which large capacity exchange system is not needed so economic advantage could be achieved.

      • 분산 시스템에서 히스토리 기법을 이용한 자기 학습 부하균등화

        구용완,정진명 수원대학교 기초과학연구소 1996 基礎科學論文集 Vol.5 No.-

        This study intended, for maximize resource utilization, to store system_state_information which periodically exchanged in distributed system into history table, and on the basic of the result, and proposed load balancing model performing nonpreemptive load balancing, and designed load balancing algorithm. When same job was scheduling later as load balancing scheduler having self-learning function on the basis of the stored content in history table(such as job type, working time, turnaround time, I/O time per job, etc., obtained from its past experience), this system positively decide whether transfer of that job to remote node is proper or not.

      • 실시간 운영체계를 위한 통신 메카니즘에 관한 연구

        丘龍完 水原大學校 1987 論文集 Vol.5 No.-

        This paper described the modification and complement of XINU operating systems to allow possible real time operating systems. The IPC machanism was developed. It is not onle optimally inserted into the communication layer of kernel for efficient synchronyzation of the processors which exist many site, but also applied to workstation based on LAN. For these purpose, the events and semaphores are introduced and designed for solving the synchronization and cooperation problems.

      • 실시간 운영 체제를 위한 사건 관리 기법

        丘龍完 수원대학교 산업기술연구소 1987 산업기술연구소논문집 Vol.2 No.-

        This paper presents an improved approach to design of real time operating system for real time event manager. This paper presents a number of refinements to basic concept that allows the real time. The two key ideas are 1) a new synchronization scheme, event, that is efficient for external event and external condition 2) a refined real time clock interrupt handler that is able to access by time counter. This paper divides the model into machine dependent part and machine independent part under the environment of IBM-PC/AT with intel 80286 microprocessor. Machine dependent part was written in assembly language and machine independent part in C language and these were implemented to apply to various application fields easily.

      • 교환기의 네트워크관리 시스템의 적체 제어에 관한 연구

        구용완,황만수 수원대학교 기초과학연구소 1992 基礎科學論文集 Vol.1 No.-

        In this paper functions were developed to control traffic and modify telephone network congestion while sensing the state of telephone lined to network where traffic congestion occurred among telephone network. TDX-10 in this system as a electronic exchange system that capacity is so large is designed to process simultaneously many calls by concurrent processing method which is different to existing exchange system that adopts sequential processing method.

      • 클러스터링을 이용한 실시간 주문형 멀티미디어 시스템에 관한 연구

        丘龍完 水原大學校 1999 論文集 Vol.17 No.-

        A MOD server is most important element in MOD system. It stores massive multimedia data in storage device such as disk array. When a MOD client requires a certain video play-back, MOD server retrieves it and transmits client. Hence, in this thesis, we analyse requirements to be considered for a MOD server and then design an efficient Soft Real-Time MOD server system using clustering.

      • 지리 정보 시스템을 위한 DXF 파일 변환기의 설계 및 구현

        문진용,구용완 수원대학교 자연과학연구소 1999 자연과학논문집 Vol.2 No.-

        In this paper, we design and implement a DXF converter which converts a DXF file into a PostScript file since the latter is one of the most powerful graphics formats and can be used for various printers directly. It consists of the File manipulation module, the DXF file manipulation module, and PostScript file generation module. In order to develop the DXF converter, we first analyze each of the file formats (i.e., DXF and PostScript formats), and then implement and test the three modules. Finally, we develop the complete DXF file converter by combining the three modules.

      • 지식형 분산 자원 관리자의 모델

        구용완 수원대학교 자연과학연구소 1998 자연과학논문집 Vol.1 No.-

        Distributed system characterized by a high degree of intercomputer resource sharing generally perform better if resources are managed utilizing as much knowledge of current global state of the system as possible. But computers in a distributed system often view the global system quite differently. Consequently, decisions which produce system-wide effects made by computers can often conflict, and causing inefficiency in resource management and therefore is to downgrade performance. This thesis presents a knowledge-based distributed resource manager to solve the related problems. It is consists of three modules. First module is a local state monitoring & updating module for monitoring events of local node. Second module is a state announcing module for distributing monitored state information throughout the distributed system. Last module is a decision making module for inference future state of each node. Also, it applicants to load balancing and performance evaluation through the simulation. Simulation result show that decrease the load imbalance and average response time of the system, also keep state to be stable average response time under surge load.

      • KCI우수등재

        실시간 처리를 위한 스케줄링 메카니즘에 관한 연구

        구용완(Yong Wan Koo),김영찬(Young Chan Kim) 한국정보과학회 1988 정보과학회논문지 Vol.15 No.3

        CPU를 공정하게 사용하기 위해 CPU 사용시간을 근거로 우선순위체제를 도입한 preemptive priority based scheduling 방법은 실시간 처리에 적합하지 못하다. 따라서 본 논문에서는 실시간 처리에 적합하도록 즉, 응답성 보장과 데드라인 처리가 가능하도록 수행시간의 긴급성에 따라(deadline) 우선순위를 부여한 preemptive priority based deadline scheduling 방법과 실시간 클럭 인터럽트로 데드라인 스케줄러를 효율적으로 지원할 수 있도록 두 개의 timer를 IBM-PC/AT의 MINIX O.S 하에서 구현했다. Preemptive priority based scheduling policy, which is based upon its CPU processing time in order to use CPU impartially, dosen't guarantee the satisfiable response time transaction. Therefore, this paper present the preemptive priority based deadline scheduling policy according to the priority of processes being determined by its urgency, so that it may be suitable for real time processes, that is, it may enable the scheduler to guarantee good response time and to processes deadline jobs. And two timers supporting deadline scheduler with real time clock interrupt efficiently were implemented in MINIX O.S under IBM-PC/AT.

      연관 검색어 추천

      이 검색어로 많이 본 자료

      활용도 높은 자료

      해외이동버튼