英語タイトル | Reduction of Traffic for Mobile Agent Migration using GAP |
---|---|
日本語タイトル | GAPを用いたキャッシュによるエージェントの同時集中移動時における通信量の削減 |
著者 | |
論文誌 | Computer Software / コンピュータソフトウェア |
巻 | 31 |
号 | 7 |
ページ | pp. 168-177 |
出版年 | 2014年7月 |
概要 | A mobile agent system is a distributed system that is constructed from software modules called a mobile agent which can work autonomously and migrate among nodes to achieve their tasks. This feature of a mobile agent is attractive to design, implement, and maintain distributed systems. A mobile agent consists of many program codes which define a task executed in each node they migrate, thus, the mobile agent migration involves the transfer of their program codes. It involves the increase of data traffic for mobile agent migrations. Several researchers proposed a mechanism using cache for a mobile agent migration to reduce data traffic of program codes. However, these mechanisms do not consider that many agents migrate at the same time, or these mechanisms increase messaging for control of transfer of program codes. Therefore, we propose a mechanism that each program code is transferred from one source node, which is assigned by the destination node when a number of mobile agents migrate at the same time. This assignment is solved as the generalized assignment problem (GAP) which is one of a problem in combinatorial optimization. Our mechanism can more reduce traffic of program codes than existing mechanisms when the number of agents, the number of program codes, the size of program codes, and the rate that each program code exists in multiple agents is great. |
ファイル | BibTeX PDF |