Fault tolerance adaptation requirements vs. quality-of-service, real-time and security in dynamic distributed systems

被引:0
作者
Tirtea, Rodica [1 ]
Deconinck, Geert [1 ]
Belmans, Ronnie [1 ]
机构
[1] Katholieke Univ Leuven, KAsteelpk Arenberg 10, B-3001 Louvain, Belgium
来源
2006 PROCEEDINGS - ANNUAL RELIABILITY AND MAINTAINABILITY SYMPOSIUM, VOLS 1 AND 2 | 2006年
关键词
adaptation; fault tolerance; quality-of-service; real-time; security; distributed systems;
D O I
10.1109/RAMS.2006.1677390
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Due to deregulation of electricity market and the trend towards distributed electricity generation based on renewable energy (e.g. wind energy), electric power infrastructure relies increasingly on communication infrastructure [I]. Communication infrastructure has to fulfil requirements as survivability and dependability to secure a high availability and reliability of electric power infrastructure. For this requirements, the communication infrastructure must be able to adapt to changes (e.g. due to dynamic environment) and to allow adaptation of quality-of-service (e.g. graceful degradation), and not the least to support adaptive fault tolerance. Given the criticality of the information transferred (e.g. control data and messages) the communication infrastructure requires support for security (e.g. confidentiality, integrity etc.) as well [2]. Because communication infrastructure handles control data, part of the traffic (or communication) has also requirements for real-time support. In this paper relations between the requirements for adaptive fault tolerance and real-time, quality-of-service and/or security are analyzed. It can be noticed that sometimes tradeoffs are required (e.g. real-time vs. fault tolerance) or the opposite, the two characteristics can use the same mechanisms to improve their support (e.g. fault tolerance vs. quality-of-service). Also, they can depend on each other for dependability support (e.g. fault tolerance vs. security). More detailed analysis depends on the specific cases (e.g. underlying distributed system, target applications such as communication infrastructure of electric power infrastructure). However, the relationships (i.e. dependencies) between the requirements for fault tolerance, real-time, quality-of-service and/or security should not be ignored, neither considered too late in the design decisions if dependability is an objective.
引用
收藏
页码:296 / +
页数:4
相关论文
共 33 条
[1]  
[Anonymous], 2001, NIST SPEC PUBL
[2]  
ATM Forum Technical Committee, 2001, ATM SEC SPEC VERS 1
[3]   Basic concepts and taxonomy of dependable and secure computing [J].
Avizienis, A ;
Laprie, JC ;
Randell, B ;
Landwehr, C .
IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING, 2004, 1 (01) :11-33
[4]  
Bao YJ, 2003, 2003 INTERNATIONAL CONFERENCE ON DEPENDABLE SYSTEMS AND NETWORKS, PROCEEDINGS, P241
[5]   Error analysis and detection procedures for a hardware implementation of the Advanced Encryption Standard [J].
Bertoni, G ;
Breveglieri, L ;
Koren, I ;
Maistri, P ;
Piuri, V .
IEEE TRANSACTIONS ON COMPUTERS, 2003, 52 (04) :492-505
[6]  
BIHAM E, 1997, P CRYPT 97
[7]   On the importance of eliminating errors in cryptographic computations [J].
Boneh, D ;
DeMillo, RA ;
Lipton, RJ .
JOURNAL OF CRYPTOLOGY, 2001, 14 (02) :101-119
[8]  
Boneh Dan, 1997, LECT NOTES COMPUTER, V1233, P37, DOI DOI 10.1007/3-540-69053-0_
[9]  
Chen TW, 1997, DIG PAP INT SYMP FAU, P21, DOI 10.1109/FTCS.1997.614073
[10]   Software-implemented fault-tolerance and separate recovery strategies enhance maintainability [J].
Deconinck, G ;
De Florio, V ;
Botti, O .
IEEE TRANSACTIONS ON RELIABILITY, 2002, 51 (02) :158-165