• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제10권1호
• /
• pp.83-93
• /
• 2006

본 논문에서는 유비쿼터스 컴퓨팅 환경에서 규칙 기반의 네트워크 장애 복구 시스템을 제안한다. 이 시스템은 유비쿼터스 컴퓨팅 환경에서 네트워크에 장애가 발생하였을 때, 정해진 규칙을 기반으로 장애를 스스로 복구할 수 있는 장애 관리 시스템이다. 시스템에 적용되는 규칙들을 네트워크 장애의 원인으로 분류하여 제시하였다. 그리고 본 논문에서는 네트워크 장애 복구 시스템이 각 상황에 맞게 규칙이 적용됨을 시뮬레이션을 통해 증명하였다.

• 제어로봇시스템학회논문지
• /
• 제4권2호
• /
• pp.264-272
• /
• 1998

This paper proposes a fault detection and recovery mechanism for a fault-tolerant Mini-MAP system, and provides detailed techniques for its implementation. This paper considers the fault-tolerant Mini-MAP system which has dual layer structure from the LLC sublayer down to the physical layer to cope with the faults of those layers. For a good fault detection, a redundant and hierarchical fault supervision architecture is proposed and its implementation technique for a stable detection operation is provided. Information for the fault location is provided from data reported with a fault detection and obtained by an additional network diagnosis. The faults are recovered by the stand-by sparing method applied for a dual network composed of two equivalent networks. A network switch mechanism is proposed to achieve a reliable and stable network function. A fault-tolerant Mini-MAP system is implemented by applying the proposed fault detection and recovery mechanism.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제2권4호
• /
• pp.281-290
• /
• 2013

IT자산은 조직의 경영목적을 지원해주는 핵심영역이며, IT자산의 장애 발생시 신속한 처리를 지원하는 것은 매우 중요하다. 본 연구에서는 IT자산의 장애가 발생할 경우, 장애해결을 위하여 기존의 장애 데이터를 기초로 장애처리 예측 기법을 제시한다. 제안한 장애처리 예측 기법은 첫째, 기존의 장애처리 데이터를 전처리하여 장애처리 유형별로 분류하고 둘째, 분류된 장애처리 유형과 장애 발생 후 접수된 내용을 키워드 매핑시키는 규칙을 제정하였으며 셋째, 제정된 규칙에 의하여 장애 발생 후 장애처리 방법이 사전에 예측 가능한 기계학습 프로세스를 제시하였다. 제시한 기계학습 프로세스의 유효성을 입증하기 위하여 A사에서 6개월 동안 접수된 33,000여건의 전산기기 장애 데이터를 실험한 결과 장애처리 예측의 적중률이 약 72%였으며, 지속적인 기계학습을 통하여 81%로 향상되었다.

• 한국정보처리학회논문지
• /
• 제7권8호
• /
• pp.2497-2504
• /
• 2000

이 논문에서는 웹 서버 장애 항목을 정의하고, 장애를 진단하기 위한 규칙 및 복구 방법을 INBANCA기법을 이용하여 제안하였다. 장애 항목으로 프로세스 장애, 서버 과부하, 인터페이스 장애, 구성 및 성능 장애를 정의하였으며, 각 장애 항목을 진단하기 위한 지식을 활성 네트워크기법을 적용하여 표현하고, 이를 시스템 레벨 장애 진단 생성규칙과 서비스 레벨에 진단 생성규칙으로 정형화하였다. 또한 사례기반의 웹 서버 장애 복구를 위한 복구 규칙 정의 및 사례 라이브러리 구축을 통하여 환경 변수에 적응적인 복구 메커니즘을 제안하였다. 그리고 제안한 장애 진단규칙 및 복구 메커니즘의 타당성을 증명하기 위한 장애 환경 구성 및 각 장애 환경에 대한 생성규칙 적용과 복구과정을 실험을 통하여 제시하였다. 이 논문에서는 기하 급수적으로 증가하는 웹 서버의 장애 관리를 위한 메커니즘을 제안함으로써 웹 서버 관리를 위한 관리자의 노력을 최소화하고, 지능적인 장애 관리를 위한 방법론을 제시하고자 한다.

• 한국전기전자재료학회논문지
• /
• 제20권9호
• /
• pp.812-815
• /
• 2007

To apply the superconducting fault current limiter(SFCL) into a power system, the analysis for its recovery characteristics as well as the consideration for its cooperation with other protecting machine such as a circuit breaker is required. The recovery characteristics of the flux-lock type SFCL like its current limiting characteristics are dependent on the winding direction of two coils. In this paper, the experiments of the current limiting and the recovery characteristics of the flux-lock type SFCL with YBCO thin film were performed. From the analysis on the experimental results due to the winding direction of two coils, the limited fault current in case of the additive polarity winding was observed to be lower than that for the case of the subtractive polarity winding. In addition, the recovery time was found to be faster in case of the additive polarity winding compared to the subtractive polarity winding.

• 정보과학회 논문지
• /
• 제44권5호
• /
• pp.460-468
• /
• 2017

실시간 클라우드의 실현에 있어서 데이터 분석 프레임워크는 중추 역할을 수행한다. 현존하는 프레임워크들 중에 가장 많은 요구사항들을 충족하는 것은 Spark Streaming이다. 하지만 이 프레임워크는 초 단위 실시간 고장 복구를 충족하지 못하고 있다. Spark Streaming의 고장 복구 기법은 정상 동작시에 기록된 누적 변형 히스토리를 토대로 고장 직전 마지막 상태 데이터를 재연산하여 복구하기 때문에 히스토리의 길이에 비례하여 복구 시간이 증가된다. 따라서 제한된 시간 이내에 고장 복구가 완료됨을 보장되지 않는다. 또한 초기 상태 데이터를 고장 감내 스토리지에서 읽는 시간이 수십 초에 달하여 초 단위고장 복구 시간을 달성할 수 없다. 본 논문에서는 언급된 문제들을 해결하기 위한 두 가지 기법들을 제안한다. 이를 Spark Streaming 1.6.2에 적용하고, 실험을 통해 고장 복구 시간이 제한 시간 이내에 완료되며 평균 약 41.57% 단축됨을 확인했다.

• 전기학회논문지P
• /
• 제59권4호
• /
• pp.477-480
• /
• 2010

Fault current in power system is expected to increase by demand of power capacity. Therefore, when the fault occurred, fault current was increased in the power system. Many studies have been progressed to limit the fault current. Superconducting fault current limiter (SFCL) is one of them which has been studied in worldwide. In this paper, we will analyze characteristics of a transformer-type SFCL by reclosing operation when the voltage increases. Twice opening times in the reclosing of circuit breaker were set as the 0.5 and 15 seconds, respectively. Turn's number of primary and secondary coils set 4:2 and we increased voltages from 120V to 280V for each experiment. By the current waveform, maximum fault current in second and third cycles was lowered when the voltage was increased. In the recovery waveform, recovery time was increased as the voltage was increased. The reason was that power burden of the SFCL increased when consumption power was increased, so the time to get back to SFCL took longer. We compared the characteristics of a resistive-type and transformer-type SFCL. As a result, we found that the fault current of a transformer-type was lower than resistive-type and recovery time of the SFCL was shorter. Consequently, transformer-type SFCL was more profitable for limitation of fault current and recovery time under the same condition for reclosing operation.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제15권2호
• /
• pp.307-311
• /
• 2015

In this paper, we propose an efficient fault-recovery technique for CGRA (Coarse-Grained Reconfigurable Architecture) based multi-core architecture. The proposed technique is intra/inter-CGRA co-reconfiguration technique based on a ring-based sharing fabric (RSF) and it enables exploiting the inherent redundancy and reconfigurability of the multi-CGRA for fault-recovery. Experimental results show that the proposed approaches achieve up to 73% fault recoverability when compared with completely connected fabric (CCF).

• International Journal of Aeronautical and Space Sciences
• /
• 제3권2호
• /
• pp.31-39
• /
• 2002

In this paper, KOMPSAT-2 on-board fault and ground recovery management design is addressesed in terms of hardware and software components which provide failure detection and spacecraft safing for anomalies which threaten spacecraft survival. It also includes ground real time up-commanding operation to recover the system safely. KOMPSAT-2 spacecraft fault and recovery management is designed such that the subsequent system configuration due to system initialization is initiated and controlled by processors. This paper will show that KOMPSAT-2 has a new design feature of CPU SEU mitigation for the possible upsets in the processor CPUs as a part of on-board fault management design. Recovery management of processor switching has two different ways: gang switching and individual switching. This paper will show that the difficulties of using multiple-processor system can be managed by proper design implementation and flight operation.

• Advances in aircraft and spacecraft science
• /
• 제4권1호
• /
• pp.53-64
• /
• 2017

Aerospace Launch Vehicles (ALV) are generally designed with high reliability to operate in complete security through fault avoidance practices. However, in spite of such precaution, fault occurring is inevitable. Hence, there is a requirement for on-board fault recovery without significant degradation in the ALV performance. The present study develops an advanced fault recovery strategy to improve the reliability of an Aerospace Launch Vehicle (ALV) navigation system. The proposed strategy contains fault detection features and can reconfigure the system against common faults in the ALV navigation system. For this purpose, fault recovery system is constructed to detect and reconfigure normal navigation faults based on the sliding mode observer (SMO) theory. In the face of pitch channel sensor failure, the original gyro faults are reconstructed using SMO theory and by correcting the faulty measurement, the pitch-rate gyroscope output is constructed to provide fault tolerant navigation solution. The novel aspect of the paper is employing SMO as an online tuning of analytical fault recovery solution against unforeseen variations due to its hardware/software property. In this regard, a nonlinear model of the ALV is simulated using specific navigation failures and the results verified the feasibility of the proposed system. Simulation results and sensitivity analysis show that the proposed techniques can produce more effective estimation results than those of the previous techniques, against sensor failures.