• Journal of Information Processing Systems
• /
• 제9권2호
• /
• pp.205-216
• /
• 2013

In this paper, we analyze a technique for building a high-availability (HA) cluster system. We propose what we have termed the 'Selective Replication Manager (SRM),' which improves the throughput performance and reduces the latency of disk devices by means of a Distributed Replicated Block Device (DRBD), which is integrated in the recent Linux Kernel (version 2.6.33 or higher) and that still provides HA and failover capabilities. The proposed technique can be applied to any disk replication and database system with little customization and with a reasonably low performance overhead. We demonstrate that this approach using SRM increases the disk replication speed and reduces latency by 17% and 7%, respectively, as compared to the existing DRBD solution. This approach represents a good effort to increase HA with a minimum amount of risk and cost in terms of commodity hardware.

• 한국정보통신학회논문지
• /
• 제20권9호
• /
• pp.1717-1722
• /
• 2016

가용도는 사용자에게 고가용 서비스를 제공하는 네트워크 시스템의 주요 성능지표로서 임의의 시간에 사용자에게 서비스를 제공할 수 있는 확률이다. 네트워크 서비스의 가용도를 향상시키는 대표적인 방법으로 다중화 방식과 재활기법이 있다. 본 논문에서는 2N redundancy와 시간 기반의 재활기법이 가용도에 미치는 영향을 분석한다. 2N redundancy는 active 시스템과 standby 시스템으로 구성하는 다중화 방식이다. 재활주기에 의존하는 시간기반의 재활기법이다. 본 논문은 시간기반의 재활기법을 적용한 2N redundnacy모델을 stochastic reward net으로 설계한다. 설계한 모델은 stochastic Petri net package를 이용하여 분석한다. 수치해석을 통해 재활기법을 적용한 시스템의 가용도가 적절한 주기에서 재활기법을 적용하지 않은 시스템의 가용도보다 높다는 것을 알 수 있다.

• ETRI Journal
• /
• 제41권4호
• /
• pp.473-482
• /
• 2019

High availability and reliability have been considered promising requirements for the support of seamless network services such as real-time video streaming, gaming, and virtual and augmented reality. Increased availability can be achieved within a local area network with the use of the virtual router redundancy protocol that utilizes backup routers to provide a backup path in the case of a master router failure. However, the network may still lose a large number of packets during a failover owing to a late failure detections and lazy responses. To achieve an efficient failover, we propose the implementation of fast detection with virtual router redundancy protocol (FDVRRP) in which the backup router quickly detects a link failure and immediately serves as the master router. We implemented the FDVRRP using open neutralized network operating system (OpenN2OS), which is an open-source-based network operating system. Based on the failover performance test of OpenN2OS, we verified that the FDVRRP exhibits a very fast failure detection and a failover with low-overhead packets.

• Journal of Information Processing Systems
• /
• 제8권2호
• /
• pp.263-278
• /
• 2012

The data grid provides geographically distributed resources for large-scale applications. It generates a large set of data. The replication of this data in several sites of the grid is an effective solution for achieving good performance. In this paper we propose an approach of dynamic replication in a hierarchical grid that takes into account crash failures in the system. The replication decision is taken based on two parameters: the availability and popularity of the data. The administrator requires a minimum rate of availability for each piece of data according to its access history in previous periods, but this availability may increase if the demand is high on this data. We also proposed a strategy to keep the desired availability respected even in case of a failure or rarity (no-popularity) of the data. The simulation results show the effectiveness of our replication strategy in terms of response time, the unavailability of requests, and availability.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 B
• /
• pp.1387-1389
• /
• 2000

Railway signalling system is required to be high safety against collision, derailment and collision at level crossing and to be high availability. The signalling system is usually divided into automatic train control, interlocking and centralized traffic control systems and each system must be high fail safe and availability. This study focused on reliability calculation of vital systems in train control system.

• 한국원자력학회:학술대회논문집
• /
• 한국원자력학회 1998년도 춘계학술발표회논문집(1)
• /
• pp.264-270
• /
• 1998

This research broadens the prime concern of nuclear power plant operations from safe performance to both economic and safe performance. First emergency diesel generator is identified as one of main contributors for the lost plant availability through the review of plants forced outage records. The framework of an integrated architecture for performing modern on-line condition for operational availability improvement is configured in this work. For the development of the comprehensive sensor networks for complex target systems, an integrated methodology incorporating a structural hierarchy, a functional hierarchy, and a fault-system matrix is formulated. The second part of our research is development of intelligent diagnosis and maintenance advisory system, which employs Bayesian Belief networks (BBNs) as a high level reasoning tool incorporating inherent uncertainty use in probabilistic inference. Our prototype diagnosis algorithms are represented explicitly through topological symbols and links between them in a causal direction. As new evidence from sensor network development is entered into the model especially, our advisory of system provides operational advice concerning both availability and safety, so that the operator is able to determine the likely modes, diagnose the system state, locate root causes, and take the most advantageous action. Thereby, this advice improves operational availability

• 정보보호학회논문지
• /
• 제24권3호
• /
• pp.547-560
• /
• 2014

스마트 그리드는 전력망에 정보통신 기술을 이용하여 에너지 이용 효율을 극대화 하는 것으로 고가용성을 요구한다. 최근 DDos공격 등 서비스중단을 통한 사회적 혼란을 야기하는 공격이 증가되고 있어 가용성에 대한 체계적인 관리가 요구된다. 한국형 스마트 그리드의 정보보호 관리체계 평가에 대해 본 논문은 가용성을 중심으로 하는 국제표준을 비교하여 새로운 평가항목을 제시하여 기존 정보보호 관리체계가 갖는 가용성 평가의 한계를 극복한다.

• 한국컴퓨터정보학회논문지
• /
• 제22권12호
• /
• pp.179-186
• /
• 2017

Managing students' data is a high prioritized duty of the university administration since most of the school affairs are proceed based on that database. Universities have invested in IT assets such as servers, storage, database, and networks. However, continuing investment in IT infrastructure is impossible due to limited budget and rapid changes in the educational environment. As cloud computing diffuse, universities are trying to reduce costs and improve efficiency by increasing server utilization, unlike when physically investing. We designed a hypothetical academic information management system based on cloud computing by utilizing the advanced server virtualization technology. This administrative cloud system allows universities to improve the availability of the system with low cost. The system demonstrates its flexibility of using data resources and immediacy of resumption.

• 한국컴퓨터정보학회논문지
• /
• 제22권4호
• /
• pp.1-8
• /
• 2017

In this paper, we propose a fault isolation system for device drivers of the Linux operating system. High availability systems impose stringent requirements upon Linux operating system. Especially device drivers can be a major source of operating system instability and many times contribute to system degradation and outages. The proposed fault isolation system identifies the occurrence of the memory-related faults in device driver and isolates it from the kernel. By operating at the early stage of the page fault handler in Linux kernel, the system detects which module causes fault and isolates it transparently from the remaining part of the kernel. By experiments, we show that the proposed system efficiently detects faults incurred by device driver, isolates the device driver and the process which accessed the driver module from the kernel.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2008년도 춘계 종합학술대회 논문집
• /
• pp.529-533
• /
• 2008

임무컴퓨터는 항공전자시스템에서 전체 시스템을 관리하고, 특정 임무를 처리하는 중요한 역할을 수행한다. 일반적으로 단일 시스템에서 SPOF(Single Point Of Failure) 요소의 고장은 전체 시스템의 고장으로 이어질 수 있으며, 이는 서비스의 중단으로 인한 임무의 실패뿐만 아니라 조종사의 생명까지도 위협할 수 있다. 이에 본 논문에서는 SPOF 요소를 제거하기 위해 단일 시스템을 이중화하여 고장발생에 유연하게 대처하도록 설계하였다. 또한 이를 효율적으로 운영하기 위한 방안으로 리눅스 기반의 Heartbeat, Fake, DRBD(Distributed Replicated Block Device), Bonding 등의 기법을 이용하여 시스템을 관리한다.