• 제어로봇시스템학회논문지
• /
• 제15권6호
• /
• pp.598-603
• /
• 2009

The checkpointing scheme is a well-known technique to cope with transient faults in digital systems. This paper proposes an adaptive checkpointing scheme for the reliability improvement of real-time control systems. The proposed adaptive checkpointing scheme is based on the previous work about the reliability problem of an equidistant checkpointing scheme. For the derivation of the adaptive scheme, some conditions are introduced which are to be satisfied for the reliability improvement by exploiting an equidistant checkpointing scheme. Numerical data show the proposed adaptive scheme outperforms the equidistant scheme from a reliability point of view.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2006년도 추계 종합학술대회 논문집
• /
• pp.609-611
• /
• 2006

분산 시스템에서 어떠한 시스템 요소에 결함이나 오류가 발생하더라도 시스템이 정상적으로 동작하게 하는 결함 허용 기법은 해당 분산시스템의 효용성을 높이는데 기여한다. 본 연구는 분산 시스템의 결함 허용 기법 중 시간 제약성 측면에서 강한 여분기반 결함허용 기법을 수용한다. 이 기법의 구성 요소인 어플리케이션 서버는 그 상태가 결정적(deterministic)이냐 비결정적(nondeterministic)이냐에 따라 그 처리 기법을 달리하고 있다. 그 중 SAR(Semi-Active Replication)이 자원 활용도 측면에서 그 효율성 증명된바 있다. 본 논문에서는 SAR의 단점인 응답시간 지연문제와 클라이언트 측면에서의 결함 허용(fault-tolerance) 문제를 해결하기 위한 기반 구조를 제안한다.

• 한국멀티미디어학회논문지
• /
• 제17권6호
• /
• pp.725-732
• /
• 2014

The Pancake graph is node symmetric and useful interconnection network in the field of data sorting algorithm. The Half Pancake graph is a new interconnection network that reduces the degree of the Pancake graph by approximately half and improves the network cost of the Pancake graph. In this paper, we analyze topological properties of the Half Pancake graph $HP_n$. Fist, we prove that $HP_n$ has maximally fault tolerance and recursive scalability. In addition, we show that in $HP_n$, there are isomorphic graphs of low-dimensional $HP_n$. Also, we propose that the Bubblesort $B_n$ can be embedded into Half Pancake $HP_n$ with dilation 5, expansion 1. These results mean that various algorithms designed for the Pancake graph and the Bubble sort graph can be executed on $HP_n$ efficiently.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2003년도 춘계학술발표논문집 (중)
• /
• pp.847-850
• /
• 2003

이동 에이전트란 컴퓨터 네트워크 상에서 사용자를 대신하여 특정 작업을 수행하는 프로그램이 독자적으로 여러 노드들을 이동하면서 필요한 작업을 수행하고 그 결과를 사용자에게 전달하도록 작성된 프로그램을 말하며, 이런 이동 에이전트를 수행할 수 있도록 컴퓨팅 환경을 제공하는 것을 이동 에이전트 컴퓨팅 환경이라고 한다. 본 연구에서는 이동 에이전트 운영에 필수 기능인 이동 에이전트 이주 기능을 이주 시작 이전에 이주 경로를 선정하여 이주 스케줄(Itinerary)로 관리하는 기법과 이동 에이전트가 이주할 시점에서 자율적으로 이주 목적지를 선정하고 선정된 목적지의 정보를 홈에 있는 Itinerary에 저장하여 관리하는 기법을 설계하였다. 그리고 이주한 방문지에서 수행 결과를 체크포인트로 저장하여 예외나 비정상적인 종료를 하였을 때 Itinerary의 내용을 보고 성공적으로 수행한 가장 마지막 플랫폼에게 이동 에이전트의 복원 및 다른 목적지로의 이주를 요청함으로서 결함 허용(fault-tolerance) 기능을 가지도록 설계하였다.

• Journal of Information Processing Systems
• /
• 제7권3호
• /
• pp.435-446
• /
• 2011

Microprocessors are becoming increasingly vulnerable to soft errors due to the current trends of semiconductor technology scaling. Traditional redundant multi-threading architectures provide perfect fault tolerance by re-executing all the computations. However, such a full re-execution technique significantly increases the verification workload on the processor resources, resulting in severe performance degradation. This paper presents a pro-active verification management approach to mitigate the verification workload to increase its performance with a minimal effect on overall reliability. An anomaly-speculation-based filter checker is proposed to guide a verification priority before the re-execution process starts. This technique is accomplished by exploiting a value similarity property, which is defined by a frequent occurrence of partially identical values. Based on the biased distribution of similarity distance measure, this paper investigates further application to exploit similar values for soft error tolerance with anomaly speculation. Extensive measurements prove that the majority of instructions produce values, which are different from the previous result value, only in a few bits. Experimental results show that the proposed scheme accelerates the processor to be 180% faster than traditional fully-fault-tolerant processor with a minimal impact on overall soft error rate.

• 전자공학회논문지A
• /
• 제33A권11호
• /
• pp.32-43
• /
• 1996

In this paper we wanted to implement a reliable ISDN PABX that could maintain the user's voice and data communication services, even when there might be an unexpected system error. In order to implement the fault-tolerance we made this ISDN PABX composed of an active and a standby side by duplicating the main control parts. Whenever a critical error occurs in the active side, it makes side exchange so that the standby side as anew active side continues to proceed the currently active voice or data services. Finally, through the performance evaluation we proved that user's voice or data communication can be serviced without any sense in spite of an system error.

• 한국시뮬레이션학회논문지
• /
• 제12권2호
• /
• pp.55-62
• /
• 2003

The explosive Increase of wireless networks and the advancement of mobile devices lead to the expansion of mobile environment. In accordance with the development of mobile environment, the need to use mobile database is increased sharply, and also it accompanies the related problems. The current mobile database system is based on the centralized method from which a synchronized server manages multiple mobile database management system to synchronize. From this mobile system architecture, several kinds of problems can be detected such as the management of synchronization issues between mobile databases and the transaction management issues. Furthermore, the current mobile database management system does not consider any solution on the fault tolerance. To solve those problems, this paper proposes the mobile agent-based mobile database management system. The proposed system provide high confidence and efficiency by enhancing the network efficiency and fault tolerance through the mobile grouping.

• 대한임베디드공학회논문지
• /
• 제9권3호
• /
• pp.183-191
• /
• 2014

The software running on avionic system is required to be highly reliable and productive. The air transport industry has developed ARINC Specification 653(ARINC653) as a standardized software requirement of avionics computers. The document specifies the interface boundary between avionics application software and the core executive software. Dependability in ARINC 653 is provided by spatial and temporal partitioning whilst fault-tolerance is provided by health monitoring mechanism. Legacy real-time operating systems are used to support ARINC653 health monitor on integrated modular avionics(IMA). However, legacy real-time operating systems are costly and difficult to modify the kernel. In this paper, we suggest a Linux-based ARINC653 health monitor. Functionalities to support ARINC653 health monitor are implemented as a Linux kernel module and its performance is evaluated.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2016년도 춘계학술발표대회
• /
• pp.1002-1003
• /
• 2016

IoT 기술은 직접적인 사람의 제어 없이 기기 간 자율 통신을 지원하며 빅데이터와 함께 미래 ICT 시장의 큰 비중을 차지할 유망 기술로써 높게 평가되고 있다. 하지만 IoT 기술 범주에 포함되는 Zigbee, Bluetooth, Bluetooth LE, Wi-Fi 등의 기술은 상호 호환되지 않고 통신 목적의 상이하기 때문에, 각 기기들은 다수의 통신 장치를 모두 탑재하고 있으며 이에 따른 운영 효율 또한 높지 않다. 또한, 통신 기술마다 존재하는 독립된 네트워크가 상호 복잡한 서비스 활용에 연계된 환경에서 특정 부분에 장애가 발생할 경우, 서비스별 관계정보 부재로 인하여 장애 발견과 복구 및 대체 서비스 제공이 어렵다. 본고는 이러한 문제를 해결하기 위하여 각 기기간의 이기종 통신I/F 연동방안과 기기 유형별로 발생될 수 있는 환경 변화에 대한 Physicality 분석을 통하여 제공 가능한 대체 서비스 분석방안을 제시한다.

• 제어로봇시스템학회논문지
• /
• 제10권8호
• /
• pp.748-754
• /
• 2004

In this paper, we propose the Triple Modular Redundancy (TMR) control system equipped with a checkpoint strategy. In this system, faults in a single processor are masked and faults in two or more processors are detected at each checkpoint time. When faults are detected, the rollback recovery is activated to recover from faults. The conventional TMR control system cannot overcome faults in two or more processors. The proposed system can effectively cope with correlated and independent faults in two or more processors. We develop a reliability model for this TMR control system under correlated and independent transient faults, and derive the reliability equation. Then we investigate the number of checkpoints that maximizes the reliability.