DocumentCode
3619688
Title
Retrieval efficiency in peer-to-peer networks with replication restrictions
Author
I.A. Klampanos;V. Poznanski
Author_Institution
Dept. of Comput. Sci., Glasgow Univ., UK
fYear
2005
fDate
6/27/1905 12:00:00 AM
Firstpage
758
Lastpage
763
Abstract
One of the most challenging problems in P2P networking is searching for relevant content efficiently. Various search architectures have been recently published, most of them addressing the problem by using distributed hash tables. However, they are hard to implement and maintain and also assume that the peers are fairly powerful computers with adequate storage and processing capabilities. In this paper, we present our experimental scenario, a photograph-sharing application, which is targeted at families and users of limited computer experience. We would expect such applications to appear, embedded in widely used home appliances, such as televisions, video players etc. Such devices cannot be expected to have the processing and input capabilities nor the memory storage capacity of modern home computers. Furthermore, acknowledging the importance of digital rights management (DRM) in multimedia sharing, we do not allow any copying, allowing peers just to view documents. We provide a search strategy for photograph albums in a P2P network based on the JXTA platform for P2P networking. We evaluate the efficiency of our network both in absolute terms and in comparison to a query-flooding alternative. We also study the effect of other parameters that would be applicable to any similar hybrid P2P network providing multimedia search services over previously annotated content.
Keywords
"Intelligent networks","Peer to peer computing","Multimedia systems","Application software","Proposals","Europe","Computer architecture","Home appliances","TV","Video sharing"
Publisher
ieee
Conference_Titel
Database and Expert Systems Applications, 2005. Proceedings. Sixteenth International Workshop on
ISSN
1529-4188
Print_ISBN
0-7695-2424-9
Type
conf
DOI
10.1109/DEXA.2005.168
Filename
1508364
Link To Document