Abstract | ||
---|---|---|
There exists a consensus that software architecture (SA) plays a central role in software development and also plays an important
role in the lifecycle phases after software delivery. Particularly, SA can be used to reduce the great difficulty and cost
of software maintenance and evolution. In this paper, runtime software architecture (RSA) based on reflective middleware is
proposed to support architecture-based software maintenance and evolution. In this approach, the actual states and behaviors
of the runtime system can be observed and manipulated in a consistent and understandable way through its architectural view.
Being an accurate, up-to-date, semantic and operable view of SA, RSA looks components and connectors as “white-box” entities
to accurately and thoroughly describe the runtime system, extends traditional architecture description languages to formally
describe itself and naturally inherit plentiful semantics in traditional views of SA, and utilizes reflective middleware to
observe and manipulate the runtime system. In order to demonstrate the feasibility of this approach, a reflective J2EE application
server, called PKUAS, is implemented to observe and manipulate the components, connectors and constraints in the runtime system.
Finally, the performance evaluation proves that making RSA explicit and operable at runtime has little effect on the runtime
system. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1360/03yf0192 | Science in China Series F: Information Sciences |
Keywords | Field | DocType |
software architecture, software maintenance, software evolution, reflective middleware | Middleware,Software architecture description,Computer science,Runtime verification,Software architecture,Reference architecture,Resource-oriented architecture,Software visualization,Distributed computing,Runtime system | Journal |
Volume | Issue | ISSN |
47 | 5 | 1862-2836 |
Citations | PageRank | References |
21 | 1.84 | 10 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gang Huang | 1 | 1223 | 110.80 |
Hong Mei | 2 | 3535 | 219.36 |
Fuqing Yang | 3 | 359 | 25.76 |