| Project/Area Number |
15200003
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (A)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Software
|
|---|
| Research Institution | National Institute of Informatics |
|---|
Principal Investigator |
HONIDEN Shinichi Information Systems Architectures Research Division, Professor, アーキテクチャ科学研究系, 教授 (70332153)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
IIJIMA Tadashia Keio University, Department of Administration Engineering Faculty of Science and Technology, lecturer, 理工学部, 講師 (20245608)
|
|---|
| Project Period (FY) |
2003 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥45,630,000 (Direct Cost: ¥35,100,000、Indirect Cost: ¥10,530,000)
Fiscal Year 2006: ¥11,050,000 (Direct Cost: ¥8,500,000、Indirect Cost: ¥2,550,000)
Fiscal Year 2005: ¥7,280,000 (Direct Cost: ¥5,600,000、Indirect Cost: ¥1,680,000)
Fiscal Year 2004: ¥7,280,000 (Direct Cost: ¥5,600,000、Indirect Cost: ¥1,680,000)
Fiscal Year 2003: ¥20,020,000 (Direct Cost: ¥15,400,000、Indirect Cost: ¥4,620,000)
|
|---|
| Keywords | agent / software development / software service / ubiquitous computing / policy / グリット / サービス / ポリシ / ミドルウェア / グリッド |
|---|
| Research Abstract |
Recent years have shown a stunning increase in the availability of hardware capable of working with an undefined number of computational nodes in order to provide services, i.e. hardware that is designed for grid computing. Software operating in a grid computing environment must not only meet the functional requirements of its provided service, but also cater to the needs of the service providers, resource providers, and end users, resulting in the need to reassess the ability of current systems to handle such requirements. Furthermore, with the impending proliferation of mobile and ubiquitous environments and the constantly changing operational conditions they imply, it will be impossible for conventional distributed software systems implemented on the base assumption of a constant environment to keep up.
Our research provides a framework to handle grid computing in a ubiquitous environment through a flexible grid agent-based architecture. First, we divided the problem space into the d
… More
etermination of design units and the logic required for implementation, of patterns describing the potential choices an agent can make, and of the policies of the service providers, resource providers and end users. We then introduced an execution environment and proposed a method for allowing the agent to choose the proper action. Lastly, we implemented a framework for grid agents and ran medium-scale tests for verification and evaluation purposes.
Specifically, we have (1) designed and constructed the supporting environment for our grid agents, (2) verified the efficacy of our methodology using medium-scale scenarios, and (3) organized and refined the specific issues involved in such an endeavor. The supporting environment runs on a cluster, is currently capable of running medium-scale scenarios involving grid agents and provides debugging facilities to aid development. We have also spent a year refining the issues and implementation while verifying the completion of our initial goals.
Through the project, we have introduced (a) conditional alterations to pre-defined policies, (b) exception handling for changes in affiliate relationships, (c) parameters for entering into parent-child relationships, changing affiliates, and individually altering the service level, and (d) customization patterns for interaction during negotiations.
We are currently working toward an internet-based domestic and international release of our completed framework. We also plan on continuing our work with the framework, applying these methods to real-world situations and ensuring the overall organization of the framework. Less
|
