| Project/Area Number |
08458062
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
計算機科学
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
MASUDA Takashi The University of Tokyo, Graduate School of Science, Professor, 大学院・理学系研究科, 教授 (80114130)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
KONO Kenji The University of Tokyo, Graduate School of Science, Research Associate, 大学院・理学系研究科, 助手 (90301118)
CHIBA Shigeru University of Tsukuba, Institute of Information Science and Electronics, Assista, 電子・情報工学系, 講師 (80282713)
|
|---|
| Project Period (FY) |
1996 – 1998
|
| Project Status |
Completed (Fiscal Year 1998)
|
| Budget Amount *help |
¥7,500,000 (Direct Cost: ¥7,500,000)
Fiscal Year 1998: ¥3,200,000 (Direct Cost: ¥3,200,000)
Fiscal Year 1997: ¥2,700,000 (Direct Cost: ¥2,700,000)
Fiscal Year 1996: ¥1,600,000 (Direct Cost: ¥1,600,000)
|
|---|
| Keywords | Reflection / Class Library / Components |
|---|
| Research Abstract |
The goal of this research is to develop a novel scheme to decompose software systems into modules by applying the reflection technique that is an object-oriented tech no logy of the next generation.
In this research, we exploited OpenC++, the language system based on reflection, which has been developed by our research group. First of all, we have extended OpenC++ to mitigate the burden of describing meta-programs that represent the relationship between modules.
The expressive power of reflection is high, but the current technology of reflection fails to decompose some classes of software systems into modules. This research developed fundamental technologies required to decompose those classes of software systems. In this research, the following software systems are addressed : fault tolerant systems, operating systems, and distributed middleware. The research on fault tolerant systems is a collaboration with LAAS in France. To decompose operating systems, we confronted the tradeoff between performance and safety, and thus we have developed a method to enable programmers to control the tradeoff.
Since OpenC++ interprets meta-programs at compile time, it is difficult to provide protection between modules during the execution of modules. This implies that modules that must be protected during execution suffer huge overhead of protection at runtime. We developed a novel mechanism of virtual memories that enable efficient protection of modules.
|
