Automated adaptations to dynamic software architectures by using autonomous agents

被引:4
作者
Jiao, WP [1 ]
Mei, H [1 ]
机构
[1] Peking Univ, Sch Elect Engn & Comp Sci, Inst Software, Beijing 100871, Peoples R China
关键词
dynamic software architectures; adaptation; goal-driven; autonomous agents; pi-calculus;
D O I
10.1016/S0952-1976(04)00108-3
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Software architectures should always reflect the architectural changes occurring in software systems. In this paper, dynamic software architectures are studied from a new perspective. Software architectures are considered as goals or objectives that software systems are pursuing instead of constraints over systems. The architectural goals are decomposed into two categories of sub-goals, i.e., structural goals and interaction behavioral goals. For defining the architectural goals of software systems, the notation of architectural reference model is introduced. The architectural reference model is composed of components and connectors involved in software architectures and constraints over software architectures. For reasoning about the result of changes rigorously and automatically, software architectures and changes to software architectures are described in a uniform formalism and the results of changes can directly be derived from the compositions of software architectures and changes. For implementing the adaptations of software systems, an autonomous agent-based approach is adopted, in which autonomous agents are used to plan the achievement of architectural goals and cascade reactions caused by changes via using architectural style-specific knowledge and behavior rules defined for agents. (C) 2004 Elsevier Ltd. All rights reserved.
引用
收藏
页码:749 / 770
页数:22
相关论文
共 37 条
[1]  
Allen R., 1997, ACM Transactions on Software Engineering and Methodology, V6, P213, DOI 10.1145/258077.258078
[2]  
Allen R, 1998, LECT NOTES COMPUT SC, V1382, P21, DOI 10.1007/BFb0053581
[3]  
ANDERWESTHUIZEN, 2002, P WORK IFIP C SOFTW, P95
[4]  
[Anonymous], IEEE COMPUTER, DOI DOI 10.1109/MC.2003.1160055
[5]   Dynamic adaptive autonomy in multi-agent systems [J].
Barber, KS ;
Goel, A ;
Martin, CE .
JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE, 2000, 12 (02) :129-147
[6]  
Brazier F. M. T., 2001, AISB Journal, V1, P5
[7]   Software architecture-based adaptation for grid computing [J].
Cheng, SW ;
Garlan, D ;
Schmerl, B ;
Steenkiste, P ;
Hu, NN .
11TH IEEE INTERNATIONAL SYMPOSIUM ON HIGH PERFORMANCE DISTRIBUTED COMPUTING, PROCEEDINGS, 2002, :389-398
[8]  
DASHOFY EM, 2002, WOSS 02, P21
[9]  
DECER K, 1997, J INTELL INF SYST, V9, P239
[10]  
DEPAULA VC, 1999, PARALLEL DISTRIBUTED, V4, P1939