UPM Institutional Repository

Agent-Based Resource Discovery In Peer-To-Peer Networks


Citation

Sedaghat, Mina (2010) Agent-Based Resource Discovery In Peer-To-Peer Networks. Masters thesis, Universiti Putra Malaysia.

Abstract

Large-scale resource sharing environments like Peer to Peers (P2P) are intrinsically distributed, heterogeneous and dynamic. Without having an efficient discovery mechanism, it is impossible to employ available resources and their related potential services that are geographically dispersed in heterogeneous platforms. Agents as an abstract encapsulated system with useful characteristics such as mobility, autonomy and intelligence can be considered as an applicable idea in different parts of P2P environments. The resource discovery model in an agent based environment can be observed from two different perspectives, in order to being improved. First, a model can be reviewed by its underlying architecture and arrangement of the network nodes. Component architecture specifies how the various components are linked together and what the components are supposed to expect from one another. It’s crucial to study the underlying architecture as it is the base platform to apply the resource discovery mechanism on top and a proper node federation will affect and facilitate the resource discovery process. The second perspective, is revisiting roles and features of the agents involved in the agent-based resource discovery model and improving their capabilities in order to achieve a higher performance.


Download File

[img]
Preview
PDF
FSKTM_2010_2_A.pdf

Download (352kB)

Additional Metadata

Item Type: Thesis (Masters)
Subject: Peer-to-peer architecture (Computer networks)
Call Number: FSKTM 2010 2
Chairman Supervisor: Professor Mohamed Othman, PhD
Divisions: Faculty of Computer Science and Information Technology
Depositing User: Mohd Nezeri Mohamad
Date Deposited: 19 Jul 2011 03:33
Last Modified: 27 May 2013 07:50
URI: http://psasir.upm.edu.my/id/eprint/11935
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item