• 제목/요약/키워드: byzantine failure

검색결과 5건 처리시간 0.022초

사이버테러를 고려한 U-Service 생존성의 정량적 평가 방안 (An Approach to a Quantitative Evaluation of U-Service Survivability Reflecting Cyber-terrorism)

  • 김성기
    • 융합보안논문지
    • /
    • 제11권6호
    • /
    • pp.67-72
    • /
    • 2011
  • 유비쿼터스 서비스(U-service)를 제공하는 시스템은 서비스 생존성이 취약한 환경을 극복해야하는 네트워크 시스템이다. 네트워크 시스템의 생존성은 시스템 구성요소에 장애나 사고, 물리적 공격이 발생하더라도 시스템에 부여된 본연의 서비스를 중단 없이 제공할 수 있는 시스템 능력으로 정의하고 있다. 본 논문에서는 비잔틴 장애를 초래하는 의도적인 사이버테러가 네트워크 시스템에 가해졌을 때의 상황을 고려하여, 사용자 입장에서 서비스 생존성을 정량적으로 평가 할 수 있는 프레임워크를 제시한다. 본 논문에서는 무선 LAN 기반의 Jini 시스템을 생존성 정량화 모델의 예로 삼는다. 그리고 Jini 시스템이 제공하는 U-service의 생존성을 평가하기 위한 연속시간 마코프 모델을 제시하고 이를 토대로 사용자가 서비스에 접근할 수 없는 확률(blocking probability)로서 U-service 생존성을 평가하는 방안을 제시한다.

블록체인 합의 방해요인 제거를 위한 Adaptive Consensus Bound PBFT 알고리즘 설계 (Adaptive Consensus Bound PBFT Algorithm Design for Eliminating Interface Factors of Blockchain Consensus)

  • 김형대;윤주식;고윤영;정종문
    • 인터넷정보학회논문지
    • /
    • 제21권1호
    • /
    • pp.17-31
    • /
    • 2020
  • 블록체인 기술이 급속도로 발전하고 있음에 따라 금융·물류 등 다양한 분야에서 블록체인 기술을 실용화하려는 시도들이 진행되고 있으며, 데이터 무결성이 매우 중요한 공공분야 또한 마찬가지이다. 국방분야 또한 네트워크 중심 작전환경(NCOE) 하에 작전운영을 위해서는 지휘통신 네트워크의 보안성 강화 및 완전무결성 확보가 매우 중요하다. 이를 위해 블록체인 네트워크를 적용한 지휘통신네트워크 구축이 필요하나, 현재까지의 블록체인 기술은 51% 공격 등의 보안 이슈들을 해결하지 못하고 있어, 국방에 접목하기 어려운 것이 현실이다. 특히, 현재 블록체인에서 많이 사용되고 있는 Practical Byzantine fault tolerance (PBFT)알고리즘은, 악의적인 행동을 하는 노드들에게 penalty 요소가 없고, 합의를 방해하는 노드가 전체 노드의 33%이상만 차지해도 합의 실패를 만드는 문제점이 있다. 본 논문에서는 블록체인의 주요 합의 알고리즘인 PBFT의 보안성 향상을 위해, Trust 모델을 접목하여 비정상 행위에 대한 penalty 메커니즘이 적용된 Adaptive Consensus Bound PBFT (ACB-PBFT) 합의 알고리즘을 제안한다.

Aseismic protection of historical structures using modern retrofitting techniques

  • Syrmakezis, C.A.;Antonopoulos, A.K.;Mavrouli, O.A.
    • Smart Structures and Systems
    • /
    • 제4권2호
    • /
    • pp.233-245
    • /
    • 2008
  • For historical masonry structures existing in the Mediterranean area, structural strengthening is of primary importance due to the continuous earthquake threat that is posed on them. Proper retrofitting of historical structures involves a thorough understanding of their structural pathology, before proceeding with any intervention measures. In this paper, a methodology is presented for the evaluation of the actual state of historical masonry structures, which can provide a useful tool for the seismic response assessment before and after the retrofitting. The methodology is mainly focused on the failure and vulnerability analysis of masonry structures using the finite element method. Using this methodology the retrofitting of historical structures with innovative techniques is investigated. The innovative technique presented here involves the exploitation of Shape Memory Alloy prestressed bars. This type of intervention is proposed because it ensures increased reversibility and minimization of interventions, in comparison with conventional retrofitting methods. In this paper, a case study is investigated for the demonstration of the proposed methodologies and techniques, which comprises a masonry Byzantine church and a masonry Cistern. Prestressed SMA alloy bars are placed into the load-bearing system of the structure. The seismic response of the non-retrofitted and the retrofitted finite element models are compared in terms of seismic energy dissipation and displacements diminution.

Seismic retrofitting of Fragavilla Monastery

  • Karantoni, Fillitsa V.
    • Earthquakes and Structures
    • /
    • 제5권2호
    • /
    • pp.207-223
    • /
    • 2013
  • Practical seismic assessment and design of retrofit for the multitude of small ecclesiastical monuments that abound in the Balkans is the subject of this work. Application of the proposed procedures and methodologies are illustrated in an example case study, a small byzantine church located in Western Greece, which is the region with the highest seismicity in Europe. The church, known as the Fragavilla Monastery, had remained almost undamaged for 800 years, until 1993 when the Pyrgos earthquake caused critical damage mainly in the vaults. Linear elastic analysis to the recorded ground motion, capped by a biaxial failure criterion reproduced the developed damage. The same modelling and analysis procedure was subsequently used for assessment of the intended retrofitting measures. Proposed retrofitting measures included mitigation of the undesirable implications of past interventions along with a combination of strengthening schemes with externally bonded AFRPs strategically placed in the structure. The effectiveness of the proposed solutions is gauged by successful reduction of stress intensity in the critical regions and mitigation of stress localization throughout the structure.

A Design for a Hyperledger Fabric Blockchain-Based Patch-Management System

  • Song, Kyoung-Tack;Kim, Shee-Ihn;Kim, Seung-Hee
    • Journal of Information Processing Systems
    • /
    • 제16권2호
    • /
    • pp.301-317
    • /
    • 2020
  • An enterprise patch-management system (PMS) typically supplies a single point of failure (SPOF) of centralization structure. However, a Blockchain system offers features of decentralization, transaction integrity, user certification, and a smart chaincode. This study proposes a Hyperledger Fabric Blockchain-based distributed patch-management system and verifies its technological feasibility through prototyping, so that all participating users can be protected from various threats. In particular, by adopting a private chain for patch file set management, it is designed as a Blockchain system that can enhance security, log management, latest status supervision and monitoring functions. In addition, it uses a Hyperledger Fabric that owns a practical Byzantine fault tolerant consensus algorithm, and implements the functions of upload patch file set, download patch file set, and audit patch file history, which are major features of PMS, as a smart contract (chaincode), and verified this operation. The distributed ledger structure of Blockchain-based PMS can be a solution for distributor and client authentication and forgery problems, SPOF problem, and distribution record reliability problem. It not only presents an alternative to dealing with central management server loads and failures, but it also provides a higher level of security and availability.