• The Journal of Information Technology and Database
• /
• v.4 no.1
• /
• pp.105-116
• /
• 1997

분산처리 시스템을 위한 소프트웨어 개발 방법 및 이를 지원하는 프레임워크의 개발은 분산처리 시스템을 위한 좋은 소프트웨어의 개발을 위해 매우 필요한 일이라 할 수 있다. 즉, 분산처리 시스템을 위한 신뢰성 있고 적합성과 확장성을 가진 소프트웨어의 개발은 매우 중요한 일이다. 자치적 분산처리 시스템(Autonomous Decentralized Systems)은 온라인 확장성과 온라인 유지보수성 및 fault tolerance기능을 가진 분산처리 시스템이다. 이 논문에서는 자치적 분산처리 시스템을 위한 객체지향 소프트웨어 개발을 지원하는 프레임워크에 대하여 논한다. 이 프레임워크는 객체지향 요구 분석과 객체지향 설계, 구현, 프로세서에 타스크(task)의 할당, 검증 및 유지보수로 구성되어 있으며, 온라인 확장성과 온라인 변경성(modifiability)을 지원한다. 자치적 분산처리 애플리케이션 소프트웨어 개발을 지원하는 프레임워크는 객체지향 computation 모델을 근간으로 하고 있다. 이 논문에서는 프레임워크외에도 자치적 분산처리 소프트웨어 개발을 위한 CASE(Computer Aided Software Engineering) 환경에 대하여서도 논한다.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.9
• /
• pp.62-69
• /
• 1996

In this paper, the bit-level 1-dimensional systolic array for modular multiplication is designed. First of all, the parallel algorithm and data dependence graph from walter's method based on montgomery algorithm suitable for array design for modular multiplication is derived. By the systematic procedure for systolic array design, four 1-dimensional systolic arrays are obtained and then are evaluated by various criteria. As it is modified the array which is derived form [0,1] projection direction by adding a control logic and it is serialized the communication paths of data A, optimal 1-dimensional systolic array is designed. It has constant I/O channels for expansile module and it is easy for fault tolerance due to unidirectional paths. It is suitable for RSA cryptosystem which deals iwth the large size and many consecutive message blocks.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.449-450
• /
• 2007

Evolvable hardware(EHW) is a dynamic field that brings together reconfigurable hardware, artificial intelligence, fault tolerance and autonomous systems. This paper gives an introduction to the field. The features that can be used to identify and classify evolvable hardware are the evolutionary algorithm, the implementation and the genotype representation. Evolvable hardware (EHW) is hardware that can change its own circuit structure by genetic learning to achieve maximum adaptation to the environment. In conventional EHW, the learning is executed by software on a computer.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.74-77
• /
• 2019

Many enterprises and public institutions are building their own server stabilization system. Nevertheless, many web servers including some public institutions have frequent obstacles due to the unexpected causes such as traffic concentration. Data loss during each of these situations and business interruption can cause a huge loss. In this study, we propose high-availability Web system that uses load balancing and session clustering. Each deals with the failure due to the over traffic by distributing incoming requests in the multiplexed web server and provides the continuous services by maintaining session information of the users in case of the failure of some servers.

• The KIPS Transactions:PartA
• /
• v.12A no.5 s.95
• /
• pp.395-404
• /
• 2005

There have been large concerns about survivability defined as the capability of a system to perform a mission-critical role, in a timely manner, in the presence of attacks, failures. In particular, One of the most important core technologies required for the design of the ITS(Intrusion Tolerance System) that performs continuously minimal essential services even when the computer system is partially compromised because of intrusions is the survivability one of In included the dependability analysis of a reliability and availability etc. quantitative dependability analysis of the In. In this Paper, we applied self-healing mechanism utilizing two factors of self-healing mechanism (fault model and system response), the core technology of autonomic computing to secure the protection power of the ITS and consisted of a state transition diagram of the ITS composed of a primary server and a backup server. We also defined the survivability, availability, and downtime cost of the ITS, and then performed studies on simulation experiments and two cases of vulnerability attack. Simulation results show that intrusion tolerance capability at the initial state is more important than coping capability at the attack state in terms of the dependability enhancement.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.9
• /
• pp.4087-4107
• /
• 2016

The current cloud computing paradigm is still vulnerable to a significant number of system failures. The increasing demand for fault tolerance and resilience in a cost-effective and device-independent manner is a primary reason for creating an effective means to address system dependability and availability concerns. This paper focuses on online failure prediction for cloud computing systems using system runtime data, which is different from traditional tolerance techniques that require an in-depth knowledge of underlying mechanisms. A 'failure prediction' approach, based on Cloud Theory (CT) and the Hidden Markov Model (HMM), is proposed that extends the HMM by training with CT. In the approach, the parameter ω is defined as the correlations between various indices and failures, taking into account multiple runtime indices in cloud computing systems. Furthermore, the approach uses multiple dimensions to describe failure prediction in detail by extending parameters of the HMM. The likelihood and membership degree computing algorithms in the CT are used, instead of traditional algorithms in HMM, to reduce computing overhead in the model training phase. Finally, the results from simulations show that the proposed approach provides very accurate results at low computational cost. It can obtain an optimal tradeoff between 'failure prediction' performance and computing overhead.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.21 no.5
• /
• pp.373-380
• /
• 2016

In this paper, a 50-kW high-efficiency modular fast charger for both electric vehicle (EV) and neighborhood electric vehicle (NEV) is proposed. The proposed fast charger consists of five 10-kW modules to achieve fault tolerance, ease of thermal management, and reduce component stress. Three-level topologies for both AC-DC and DC-DC converters are employed to use 600V MOSFET, resulting in ease of component selection and increase in switching frequency. The proposed three-level DC-DC converter with coupled inductor and its hybrid switching method can reduce the circulating current under wide output voltage range. A 50-kW prototype of the proposed fast charger was developed and tested to verify the validity of the proposed concept. Experimental results show that the proposed fast charger achieves a rated efficiency of 95.2% and a THD of less than 3%.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.5
• /
• pp.1330-1340
• /
• 2000

Multimedia isn ow applied to various real world areas. In particular, the focus on multimedia system and CSCW(Computer Supported Cooperative Work) has increased. In spite of this current trend, however, the study of fault tolerance for CSCW has not yet fully progressed. We propose EDR_MSCW. It is a system that is suitable for detecting ad recovering software error based on multimedia computer supported cooperative work as DOORAE by using software techniques. DOORAE is a framwork for supporting development on multimedia applications for computer-based collaborative works. When an error occurs, EDR_MCSCW detects an error by using hooking methods in MS-Windows API(Application Program Interface) function. If an error is found, we present a checkpointing and recovery algorithm which has the removal function of the domino-effect for recovering multimedia and CSCW by using stack.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.4A
• /
• pp.304-311
• /
• 2005

OCSP has specific characters which can suspend, close, and correct in real time. But, as more clients use the OCSP server verification, more updated information is needed, which can lead to jamming in the OCSP server. To apply this technique of Distributed OCSP server so as to reduce the certificate verification OCSP from jamming. Also, the Distributed OCSP server will solve the problems of the intensive central structure. Simulation results show that the average reply time of certificate verification request and server load are reduced in the case using distributed OCSP. In addition to this advantage, resource distribution and fault tolerance are acquired because of multiple OCSP.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.9
• /
• pp.3126-3145
• /
• 2014

Cloud computing is a computing paradigm in which users can rent computing resources from service providers according to their requirements. A spot instance in cloud computing helps a user to obtain resources at a lower cost. However, a crucial weakness of spot instances is that the resources can be unreliable anytime due to the fluctuation of instance prices, resulting in increasing the failure time of users' job. In this paper, we propose a Genetic Algorithm (GA)-based workflow scheduling scheme that can find the optimal task size of each instance in a spot instance-based cloud computing environment without increasing users' budgets. Our scheme reduces total task execution time even if an out-of-bid situation occurs in an instance. The simulation results, based on a before-and-after GA comparison, reveal that our scheme achieves performance improvements in terms of reducing the task execution time on average by 7.06%. Additionally, the cost in our scheme is similar to that when GA is not applied. Therefore, our scheme can achieve better performance than the existing scheme, by optimizing the task size allocated to each available instance throughout the evolutionary process of GA.