• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.1276-1279
• /
• 2008

최근 모바일 장치의 증가와 무선 네트워크 환경의 발전은 모바일 애드혹 네트워크에 대한 관심을 증가시키고 있다. 최근까지의 연구는 모바일 애드혹 네트워크에서 데이터 접근성(Accessibility) 향상을 위해 복제의 활용이 효율적임을 증명하고 있다. 복제 기법이 모바일 애드혹 네트워크에서 실제로 활용되기 위해서는 원본과 복제본 사이에 일관성(Consistency)이 확보되어야만 한다. 기존의 연구에서는 일관성 확보를 위해 원본을 가진 노드에게 확인하는 기법을 활용하고 있다. 그러나, 시간제약을 가지는 응용의 경우 원본을 가진 노드에게 확인하는 기법은 효율적이지 못하며, 때로는 심각한 문제를 일으킬 수도 있다. 본 논문에서는 시간제약응용을 위한 캐쉬 무효화 기법을 제안하고 있다. 제안하는 기법을 통해 기존의 일관성 확인 기법에서 일어나는 '롤백(Rollback)' 문제를 해결할 수 있었으며, 접근시간(Access time) 측면에서 성능향상을 얻을 수 있었다. 실험 결과는 본 논문의 접근시간 측면에서의 효율적임을 증명하여 준다.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.11
• /
• pp.1074-1081
• /
• 2011

As robotics systems are becoming more complex there is the need to promote component based robot development, where systems can be constructed as the composition and integration of reusable building block. One of the most important challenges facing component based robot development is safeguarding against software component failures and malfunctions. The health monitoring of the robot software is most fundamental factors not only to manage system at runtime but also to analysis information of software component in design phase of the robot application. And also as a lot of monitoring events are occurred during the execution of the robot software components, a simple data treatment and efficient memory management method is required. In this paper, we propose an efficient events monitoring and data management method by modeling robot software component and monitoring factors based on robot software framework. The monitoring factors, such as component execution runtime exception, Input/Output data, execution time, checkpoint-rollback are deduced and the detail monitoring events are defined. Furthermore, we define event record and monitor record pool suitable for robot software components and propose a efficient data management method. To verify the effectiveness and usefulness of the proposed approach, a monitoring module and user interface has been implemented using OPRoS robot software framework. The proposed monitoring module can be used as monitoring tool to analysis the software components in robot design phase and plugged into self-healing system to monitor the system health status at runtime in robot systems.

• Journal of KIISE:Information Networking
• /
• v.29 no.5
• /
• pp.562-572
• /
• 2002

Distributed object technologies have become popular in developing distributed systems. Although such middleware platforms as DSOM, DCOM, CORBA and Java RMI ease the development of distributed applications, they do not directly improve the reliability and the availability of these applications. Because the task of developing fault-tolerance techniques for distributed object paradigms is often complicated and error-prone, there is a great need for a development toolkit that enhances the reliability and the availability of distributed objects. In this paper, we propose a fault-tolerant distributed middleware system based on RMI, called IMMORTAL. We use a log-based rollback-recovery mechanism for supporting reliable distributed computing. Through a series of experiments, we observe that benchmark applications on the IMMORTAL tolerate hardware and software failures and evaluate its performance and scalability.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.3
• /
• pp.202-207
• /
• 2016

Checkpoint placement is an effective fault tolerance technique against transient faults in which the task is re-executed from the latest checkpoint when a fault is detected. In this paper, we propose a new checkpoint placement strategy separating data saving and fault detection processes that are performed together in conventional checkpoints. Several fault detection processes are performed in one checkpoint interval in order to decrease the latency between the occurrence and detection of faults. We address the placement method of fault detection processes to maximize the probability of successful execution of a task within the given deadline. We develop the Markov chain model for a real-time task having the proposed checkpoints, and derive the optimal fault detection and checkpoint interval.

• The KIPS Transactions:PartA
• /
• v.10A no.2
• /
• pp.123-130
• /
• 2003

This paper presents a new striped checkpointing scheme for serverless cluster computers, where the local disks are attached to the cluster nodes collectively form a distributed RAID with a single I/O space. Striping enables parallel I/O on the distributed disks and staggering avoids network bottleneck in the distributed RAID. We demonstrate how to reduce the checkpointing overhead and increase the availability by striping and staggering dynamically for communication intensive applications. Linpack HPC Benchamark and MPI programs are applied to these checkpointing schemes for performance evaluation on the 16-nodes cluster system. Benchmark results prove the benefits of the striped checkpointing scheme compare to the existing schemes, and these results are useful to design the efficient checkpointing scheme for fast rollback recovery from any single node failure in a cluster system.

• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.5_6
• /
• pp.328-334
• /
• 2004

Recently, Software DSMs continue to improve its performance and scalability As Software DSMs become attractive on larger clusters, the focus of attention is likely to move toward improving the reliability of a system. A popular approach to tolerate failures is message logging with checkpointing, and so many log-based rollback recovery schemes have been proposed. In this work, we propose a remote logging scheme which uses the volatile memory of a remote node assigned to each node. As our remote logging does not incur frequent disk accesses during failure-free execution, its logging overhead is not significant especially over high-speed communication network. The remote logging tolerates multiple failures if the backup nodes of failed nodes are alive. It makes the reliability of DSMs grow much higher. We have designed and implemented the FT-KDSM(Fault Tolerant KAIST DSM) with the remote logging and showed the logging overhead and the recovery time.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.7
• /
• pp.767-777
• /
• 1999

인과적 메시지 로깅 기법은 정상프로세스를 역전(roll-back)시키거나 메시지의 저장을 위해 프로세스의 수행을 중단시키지 않는 장점을 지니고 있지만, 메시지의 크기가 지나치게 커진다는 단점을 지니고 있다. 본 논문에서는 인과적 메시지 로깅 기법의 이러한 문제점을 해결하기 위하여 로그 상속의 개념을 정의하고 로그 연혁을 이용하여 로그 비용, 특히 로그 크기 면에서 효율적인 로깅 기법을 제안한다. 또한 이 로깅 알고리즘을 이용하여 복구시 메시지의 수와 크기를 줄여 복구시간을 줄이는 효율적인 복구 알고리즘을 제안하고, 제안한 알고리즘이 메시지 로그 크기 면에서 효율적임을 증명한다. 또 제안한 알고리즘의 성능을 검증하기 위하여 두 가지 종류의 모의 실험을 수행하여 기존의 로깅 프로토콜과 메시지 크기 면에서의 성능을 비교한 결과를 제시하였다.Abstract Causal message logging has many good properties such as nonblocking message logging and no rollback propagation. However, it requires a large amount of information to be piggybacked on each message, which may incur severe performance degradation. This paper presents an efficient causal logging algorithm based on the new message log structure, LogOn, which represents the causal inter-process dependency relation with much smaller overhead compared to the existing algorithms. The proposed algorithm is efficient in the sense that it entails no additional information other than LogOn to be carried in each message, while other existing algorithms require extra information other than the message logs. This paper also presents an efficient recovery algorithm to solve the problem of a large amount of data exchanges during the recovery. To verify the performance of our algorithm, we give an analysis of the algorithm and perform two simulations and compare the log size with other causal logging protocols.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.3
• /
• pp.527-534
• /
• 2023

Triple modular redundancy (TMR) systems can continue their mission by virtue of their structural redundancy even if one processor is attacked by faults. In this paper, we propose a new fault tolerance strategy by introducing checkpoints into the TMR system in which data saving and fault detection processes are separated while they corporate together in the conventional checkpoints. Faults in one processor are tolerated by synchronizing the state of three processors upon detecting faults. Simultaneous faults occurring to more than one processor are tolerated by re-executing the task from the latest checkpoint. We propose the checkpoint placement and fault detection strategy to maximize the probability of successful execution of a task within the given deadline. We develop the Markov chain model for the TMR system having the proposed checkpoint strategy, and derive the optimal fault detection and checkpoint interval.

• The Korean Journal of Petroleum Geology
• /
• v.13 no.1
• /
• pp.1-16
• /
• 2007

The southeastern Korean Peninsula has experienced crustal multi-deformations according to changes of global tectonic setting during the Cenozoic. Characteristic features of the crustal deformations in relation to major Cenozoic tectonic events are summarized as follows. (1) Collision of Indian and Eurasian continents and abrupt change of movement direction of the Pacific plate (50${\sim}$43 Ma): The collision of Indian and Eurasian continents caused the eastward extrusion of East Asia block as a trench-rollback, and then the movement direction of the Pacific plate was abruptly changed from NNW to WNW. As a result, the strong suction-force along the plate boundary produced a tensional stress field trending EW or WNW-ESE in southeastern Korea, which resultantly induced the passive intrusion of NS or NNE trending mafic dike swarm. (2) Opening of the East Sea (25${\sim}$16 Ma): The NS or NNW-SSE trending opening of the East Sea generated a dextral shear stress regime trending NNW-SSE along the eastern coast line of the Korean Peninsula. As a result, pull-apart basins were developed in right bending and overstepping parts along major dextral strike slip faults trending NNW-SSE in southeastern Korea. The basins can be divided into two types on the basis of geometry and kinematics: Parallelogram-shaped basin (rhombochasm) and wedged-shaped basin (sphenochasm), respectively. In those times, the basins and adjacent basement blocks experienced clockwise rotation and northwestward tilting contemporaneously, and the basins often experienced a kind of propagating rifting from NE toward SE. At about 17Ma, the Yonil Tectonic Line, which is the westernmost border fault of the Miocene crustal deformation in southeastern Korea, began to move as a major dextral strike slip fault. (3) Clockwise rotation of southeastern Japan Island (about 15 Ma): The collision of the Izu-Bonin Arc and southeastern Japan Island, as a result of northward movement of the Philippine sea-plate, induced the clockwise rotation of southeastern Japan Island. The event caused the NW-SE compression in the Korea Strait as a tectonic inversion, which resultantly tenninated the basin extension and caused local counterclockwise rotation of blocks in southeastern Korea. (4) E-W compression in the East Asia (after about 5 Ma): Decreasing subduction angle of the Pacific plate and eastward movement of the Amurian plate have constructed the-top-to-west thrusts and become a major cause for earthquakes in southeastern Korea until the present time.

• The Journal of the Petrological Society of Korea
• /
• v.16 no.2 s.48
• /
• pp.82-99
• /
• 2007

The Paleogene dikes intruding into the late Cretaceous granodiorite are pervasively observed in the Irun-myeon, eastern Geoje Island. They are classified into three groups: NW-trending acidic dike swarm and WNW- (A-Group) and $NS{\sim}NNE-trending$ (B-Group) basic dike swarms. Based on their cross-cutting relationships, the earliest is the acidic dike group and fellowed by A- and B-Groups in succession. The acidic dikes seem to have intruded into tension gashes induced by the sinistral strike-slip faulting of the Yangsan fault system during the late $Cretaceous{\sim}early$ Paleogene. In terms of rock-type, orientation, age, and geochemistry, A-Group and B-Group are intimately correlated with the intermediate and basic dike swarms in the Gyeongju-Gampo area, respectively. These results significantly suggest that the corresponding dike swarms are genetically related. Based on the K-Ar and Ar-Ar age data, A- and B- Groups were intruded during $64{\sim}52\;Ma$ and $51{\sim}44\;Ma$, respectively. The result means that the direction of tensional stress in and around the SE Korean peninsula was changed abruptly from NNE-SSW to $EW{\sim}WNW-ESE$ at about 51 Ma. Considering the tectonic environments during the Paleogene, it is interpreted that A-Group was injected along the WNW-trending tensional fractures developed under an regional sinistral simple shear regime which was caused by the north-northwestward oblique subduction of the Pacific plate beneath the Eurasian plate. Meanwhile, the regional stress caused by the collision of India and Eurasia continents at about 55 Ma was likely propagated to the East Asia at about 51 Ma, and then the East Asia including the Korean peninsula was extruded eastwards as a trench-rollback and the dip of downgoing slab of the Pacific plate was abruptly steepened. As a result, the strong suction-force along the plate boundary produced a tensional stress field trending EW or WNW-ESE in and around the Korean peninsula, which resultantly induced B-Group to intrude passively into the study area.