• Journal of the Korea Society for Simulation
• /
• v.1 no.1
• /
• pp.64-76
• /
• 1992

The DEVS(discrete event system specificaition) formalism specifies a discrete event system in a hierarchical, modular form. DEVSIM++ is a C++based general purpose DEVS abstract simulator which can simulate systems modeled by the DEVS formalism in a sequential environment. This paper describes P-DEVSIM++which is a parallel version of DEVSIM++ . In P-DEVSIM++, the external and internal event of DEVS models can by processed in parallel. For such processing, we propose a parallel, distributed optimistic simulation algorithm based on the Time Warp approach. However, the proposed algorithm localizes the rollback of a model within itself, not possible in the standard Time Warp approach. An advantage of such localization is that the simulation time may be reduced. To evaluate its performance, we simulate a single bus multiprocessor architecture system with an external common memory. Simulation result shows that significant speedup is made possible with our algorithm in a parallel environment.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.04a
• /
• pp.1134-1136
• /
• 2012

SQLite는 임베디드 환경에 최적화된 경량형 로컬 데이터베이스 시스템으로서, 대부분의 스마트폰을 비롯한 임베디드 시스템과 사용자 애플리케이션에 라이브러리로 사용되고 있다. SQLite는 데이터베이스 시스템의 중요한 특성인 Atomic Write를 보장하기 위하여 롤백 저널(Rollback Journal)과 WAL(Write-Ahead Log)의 두 가지 저널링 알고리즘 가운데 하나를 선택하여 사용한다. 본 연구에서는 이 두 알고리즘의 동작 원리를 분석하고 동작 성능을 측정하여 그 원인을 분석하고 개선 가능성을 보인다.

• Journal of Advanced Information Technology and Convergence
• /
• v.10 no.2
• /
• pp.83-90
• /
• 2020

A protocol HMNR, was proposed to utilize control information of every other process piggybacked on each sent message for minimizing the number of forced checkpoints. Then, an improved protocol, called Lazy-HMNR, was presented to lower the possibility of taking forced checkpoints incurred by the asymmetry between checkpointing frequencies of processes. Despite these two different minimization techniques, if the high message interaction traffic occurs, Lazy-HMNR may considerably lower the probability of knowing whether there occurs no Z-cycle due to its shortcomings. Also, we recognize that no previous work has smart procedures to be able to utilize network infrastructures for highly decreasing the number of forced checkpoints with dependency information carried on every application message. We introduce a novel Lazy-HMNR protocol for group communication-based distributed computing systems to cut back the number of forced checkpoints in a more effective manner. Our simulation outcomes showed that the proposed protocol may highly lessen the frequency of forced checkpoints by comparison to Lazy-HMNR.

• ETRI Journal
• /
• v.46 no.1
• /
• pp.59-70
• /
• 2024

We introduce a high-performance named entity recognition (NER) model for written and spoken language. To overcome challenges related to labeled data scarcity and domain shifts, we use transfer learning to leverage our previously developed KorBERT as the base model. We also adopt a meta-pseudo-label method using a teacher/student framework with labeled and unlabeled data. Our model presents two modifications. First, the student model is updated with an average loss from both human- and pseudo-labeled data. Second, the influence of noisy pseudo-labeled data is mitigated by considering feedback scores and updating the teacher model only when below a threshold (0.0005). We achieve the target NER performance in the spoken language domain and improve that in the written language domain by proposing a straightforward rollback method that reverts to the best model based on scarce human-labeled data. Further improvement is achieved by adjusting the label vector weights in the named entity dictionary.

• The KIPS Transactions:PartC
• /
• v.16C no.1
• /
• pp.65-72
• /
• 2009

Due to the popularity of mobile devices and advances in wireless communication technologies, a mobile ad hoc network has received a lot of attention. In the existing data replication management research, the use of a replica has been shown to be an efficient technique for improving data accessibility. However, to use a replica in ad hoc networks, the data consistency between the original data and the replica should be guaranteed. In the traditional research, a mobile node should check an original data whether the data is updated or not. However, It may be costly or sometimes impossible to check the original data. In the case of the time constraint applications, the checking cost can cause more serious problem. In this paper, we propose the time-guarantee based cache invalidation schemes for time constraint applications and the threshold based compensation method to enhance the time-guarantee based scheme. The proposed schemes can remove the "rollback" problem. Simulation results show that our schemes outperform the previous ones in terms of access time with little loss of data currency.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.11
• /
• pp.127-133
• /
• 2019

SQL is the most common language in data processing. Therefore, most of the colleges offer SQL in their curriculum. In this research, an auto scoring SQL test is proposed for the efficient results of SQL education. The system was treated with algorithms instead of using expensive DBMS(Data Base Management System) for automatic scoring, and satisfactory results were produced. For this system, the test question bank was established out of 'personnel management' and 'academic management'. It provides users with different sets of test each time. Scoring was done by dividing tables into two sections. The one that does not change the table(select) and the other that actually changes the table(update, insert, delete). In the case of a search, the answer and response were executed at first and then the results were compared and processed, the user's answers are evaluated by comparing the table with the correct answer. Modification, insertion, and deletion of table actually changes the data table, so data was restored by using ROLLBACK command. This system was implemented and tested 772 times on the 88 students in Computer Information Division of our college. The results of the implementation show that the average scoring time for a test consisting of 10 questions is 0.052 seconds, and the performance of this system is distinguished considering that multiple responses cannot be processed at the same time by a human grader, we want to develop a problem system that takes into account the difficulty of the problem into account near future.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.768-770
• /
• 1998

동기적 검사점(synchronous checkpoiting)기법, 인과적 메시지 로깅(causal message logging)과 향상된 회복 비동기성(improved asynchronism during recovery)을 제공하는 복귀회복(rollback recovery) 기법을 적용하여 자바 메시지 전달 시스템(java massage passing system)에서 수행하는 병렬 에플리케이션들에게 저 비용의 결함 포용성에 따라, 통신망으로 연결된 이질형 (fault-tolerance)(heterogeneous) 컴퓨터들을 이용하는 대규모 분산 시스템들은 아주 효율적인 병렬 컴퓨팅 환경을 제공해준다. 그러나, 이러한 분산 시스템들의 규모가 커짐에 따라 고장률 (failure rate)도 그 만큼 중요하게 된다. 따라서, 고장률이 높은 대규모 분산 시스템들에게 좀더 효율적인 결함 포용성을 제공하는 기법들이 필요하다. 또한, 대규모분산 시스템들은 이질형 컴퓨터들로 구성되어 있기 때문에, 결함 포용성을 제공하는 소프트웨어 패키지들은 플랫폼 독립적(platform independent)이어야 한다. 이러한 문제점은 높은 이식성(portability)을 가지고 있는 자바 언어로 구현함으로써 해결될 수 있다. 따라서, 본 논문은 자바 메시지 전달 시스템에서 수행되는 병렬 애플리케이션들에게 동기적 검사점 기법, 인과적 메시지 로깅과 향상된 비동기성을 제공하는 복귀회복 기법을 높은 이식성을 가진 자바언어로 구현하여 저 비용으로 결함 포용성을 제공하고자 한다.

• The Journal of Korean Association of Computer Education
• /
• v.7 no.3
• /
• pp.35-43
• /
• 2004

Internet-based P2P computing with independent machines suffers from frequent disconnections and security threats caused by leaving, failure, network diversity, or anonymity of participated machines. Replication schemes of shared resources are used for solving these issues in many studies and implementations. We propose an operational replication scheme in P2P computing to share computing resources, and the scheme verifies the correctness of operation against faults and security threats. This verifications are carried out periodically on replicated and dependent working units without global message exchanges over the whole system. The verified working units are treated as checkpoints, and thus they could be put to practical use for fault-tolerance with rollback recovery.

• Proceedings of the KSR Conference
• /
• 2009.05a
• /
• pp.831-839
• /
• 2009

An interlocking equipment of railway signalling systems is manufactured with electrical devices and electrical interlocking equipment has been substituted for existing interlock equipment(relay sequential logic type). But interlock conditions are still generated from rail diagram and used to make Interlock Table manually. In order to make EIS(Electrical Interlock System) operate, we should write interlock data which is made from interlock table. But, as the station is larger and more complicated, handwork may becomes a very tediou work and makes more mistakes. Therefore the development of CAD system for Interlocking System is very significant, if CAD can reduce the mistakes from handwork and help the configuring the interlocking system. In this paper, we first arrange some rules which can be used to extract route information automatically from rail diagram and interlocking rules. And then we propose "Search-And-Rollback" algorithm to extract route information and individual interlocking rules. The proposed algorithm is implemented and tested through the signal design process of the Hyundai-Steel private railway to carry melted pig iron from the blast furnace to the steel-making workshop. some cases. It shows that CAD for Interlocking system is very helpful in time saving aspect and system reliability.

• The KIPS Transactions:PartA
• /
• v.18A no.5
• /
• pp.177-180
• /
• 2011

In this paper, we present a record-replay technique based on interrupt logging and reproduction. Race conditions have been considered as the main source of nondeterminism in conventional record-replay approaches. However, interrupts are another source of nondeterministic computer system behavior, which must be reproduced at accurate time points, let alone the order of interrupt occurrence. We show that an interrupt-based replayer can be efficiently and effectively implemented by using hardware performance counters and debugging functionality. Experiments also show that the runtime overhead of the interrupt-based replayer is sufficiently low.