| dc.description.abstract | Multi-agent systems are capable of handling the complexity of the real world problems through its emerging features including coordination, communication and negotiations. Thus, Multi-agent systems can be applied to handle the environmental complexity of a computer network to achieve better performance and reduce resource wastage. For instance, some of the clients contain the huge workload when other clients are free on the network. In this situation, some task can be allocated to free clients to complete all tasks as much as possible. Further, most of the network resources are wasting for download same file in many clients. This resource wastage can be reduced by checking each client before downloading the file. This paper presents a distributed multi-agent system that can be used to reduce resource wastage of the Local Area Network. The Multi-agent system has been implemented through the MaSMT framework with selections of a managing agent and four ordinary agents, namely, file sendagent, file receive agent, download agent and load balancing and dynamic scheduling agent. When the user feeds an URL to the system for download, the system communicates with other manager agents in the Local Area Network and checks whether the file is available or not. If the file is available, get that file from that client. Otherwise, the system will download the file through the file download agent. When the user feeds several URLs, the system follows previous procedure and allocates downloading tasks to other clients through the load balancing and dynamic scheduling agent when they are free. Task allocation, file sending and reserving has been done, though the agent-agent communication. The system has been successfully tested in a laboratory environment and successful results were archived when downloading numbers of large files such as video playlist. | en_US |
