Shinji Sugawara

A utility based contents sharing method in hybrid peer-to-peer networks with cloud storages

Recently, various contents sharing methods that are available on peer-to-peer networks have been actively researched; a peer-to-peer network generally consists of a number of terminals or computer systems evenly connected to each other. In most of previous studies, shared content items are supposed to be kept in the storages of the peers, and effective ways of contents sharing under the environment where not only the storages of peers but also ones of cloud systems are available have not sufficiently researched yet. Accordingly in this paper, we propose an efficient contents sharing method in a hybrid peer-to-peer network under the environment where storages of both peers and cloud system are available, and then verify the effectiveness of the proposed method by computer simulations.

An Efficient Content Searching Method Using Query Transmission Records with Content Cache Routers in Unstructured Peer-to-Peer Networks

Various methods have been investigated for content sharing in peer-to-peer (P2P) networks that are evenly connected to each other. However, the network load especially in unstructured P2P networks increases due to redundant generations of queries when a flooding-based content searching method is adopted because no one can control the location information of each content item in the network. Here, we propose an efficient content searching method for unstructured P2P networks, where each of peer and content cache router keeps query transmission records, which are used to guide content searching queries. In addition, we discuss the effectiveness of our proposal using computer simulations.

An Efficient Content Searching Method Using Transmission Records with Wasted Queries Reduction Scheme in Unstructured Peer-to-Peer Networks

A lot of content searching methods applied to Peer-to-Peer (P2P) networks have been researched recently. However, especially in unstructured P2P networks, the amount of traffic grows when a flooding-based content searching is adopted because of a lot of wasted queries. Accordingly, we proposed a content searching method for unstructured P2P networks in our former research, in which each peer keeps query transmission records and guides queries properly so as to reduce wasted ones according to the records by canceling query transmissions to the peers with low content existing probability. In this paper, we improve the proposed method by adding wasted query reduction scheme which works while the usual flooding is executed, and reduce the network traffic without degradation of content acquisition rate. In addition, we discuss the availability of the newly proposed method with its evaluation results given by computer simulations.

Efficient Query Transmission Using Route Selection Records and Multiple Dynamic Network Agents

Various query delivery methods in unstructured Peer-to-Peer networks have been researched recently. However, it is not clear whether these methods are effective under the condition where some peers are joining in and dropping out of a network or some contents newly appear and the other disappear frequently. Therefore in this paper, we propose a query guidance method using route selection records and multiple dynamic network agents. The effectiveness of the proposed method is evaluated by computer simulations.

Efficient Content Sharing with File Splitting and Differences Between Versions in Hybrid Peer-to-Peer Networks

This paper proposes an efficient content sharing strategy using file splitting and difference between versions in hybrid Peer-to-Peer (P2P) networks. In this strategy, when a user requests a content item, he/she can get it from the network by retrieving the other version of the content item and the difference from the requested version, if the obtaining cost of the requested version is expensive. This way of content sharing can be expected to accomplish effective and flexible operation. Furthermore, efficient utilization of a peer’s storage capacity is achieved by splitting each replica of a content item into several small blocks and storing them separately in the plural peers.

Autonomous Decentralized System for Knowledge Refinement of Contents Published over Networks

Recently, a large number of useful contents are accumulated in large-scale networks like the Internet. On the other hand, plural sources often provide slightly different content concerning a same knowledge, and this causes incoherence of the knowledge for the network users. This paper proposes an autonomous decentralized knowledge refinement system that provides reliable knowledge to the users. This system consists of plural subsystems, and they autonomously subscribe contents from a lot of content sources over the network, refine and reconstruct the contents, and publish reliable knowledge. The effectiveness of this system is evaluated by computer simulations.

An expected acquisition rate based contents sharing scheme in hybrid P2P networks and cloud storages

This paper makes a proposal of efficient contents sharing scheme working under the situation where the storages of both hybrid Peer-to-Peer (P2P) networks and a cloud system are available. The scheme keeps the expected acquisition rate for contents searching at the level of freely-prearranged value, and is able to provide ideal environments according to the different types of contents sharing service requirements. The effectiveness of the scheme is verified by computer simulations.

A Content Searching Scheme Using Popularity and Link Degree of Nodes in Unstructured P2P Networks with Cache Routers

This paper improves the conventional scheme of contents sharing in unstructured P2P networks, where each of the peers and content cache routers keeps contents cache as well as query transmission records, which are used to guide content searching queries and reduce the number of the queries without losing contents discovery rates. In this paper, the decision whether a content item is cached or not on the router is executed as a part of the algorithm running in the contents sharing system according to the content items popularity and the degree of the router as a network node, in order to reduce the frequency of contents loss caused by the displacement or overwriting of caches in the routers. The efficiency of the improvement is discussed using the results of computer simulations.

An efficient digital content sharing considering regional variety of popularities in hybrid peer-to-peer networks with cloud storages

A variety of contents sharing methods that are available on Peer-to-Peer networks have been actively researched these days. We proposed a system using not only the peers storages but also cloud storages for possessing the shared content items, which achieved higher contents acquisition rate and lower network load than conventional systems. However in the research, the situation where the popular content items are generally different among nations or regions was not considered. Therefore this research proposes the efficient contents sharing method taking into account the difference of contents popularities depending on regions, and in addition, evaluates the efficiency of the proposed method by computer simulations.

An efficient digital contents sharing with transmission records in peer-to-peer networks

For unstructured peer-to-peer (P2P) networks, we previously proposed an efficient digital content sharing method, in which content searching queries are guided properly to an appropriate peer which is likely to store a replica of target content item, by using query transmission records (TRs) maintained in each peer. In this paper, we improve the conventional method and make it achieve more efficiency especially in the situation where some content items which are likely to be highly interesting to some users are released at short intervals.