• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10e
• /
• pp.460-462
• /
• 2002

컴퓨터 네트워킹의 새로운 패러다임으로 떠오르고 있는 P2P(Peer-to-peer)통신은 기존의 서버-클라이언트 모델의 한계점을 극복하는 많은 해법을 제시하고 있다. 이에 최근여러 가지 플랫폼의 각기의 서비스를 제공하는 애플리케이션들이 속속 등장하고 있다. SunMicro systems에서 시작되어 현재 Open Project로 진행중인 Project JXTA는 이러한 분산 Computing, 특히 Peer-to-Peer Computing이나 Peer-to-Peer 네트워킹이라는 영역에서 폭넓게 발생하는 문제를 해결하기 위해 만들어진 테크놀로지로서 Peer to Peer를 비롯한 분산 Computing들을 위한 Infra Service를 제공하고 있다. JXTA 테크놀로지를 이용하여 제공할 수 있는 여러 서비스 중에 일반 Peer들의 대용량 데이터 전송에서 있을 수 있는 문제점을 해결할 수 있는 서비스를 제안한다. 대용량의 데이터를 여러 서버에서 나누어 받을 수 있도록 하여 Peer-to-Peer 어플리케이션에서 사용자는 찾고자 하는 자료를 검색했을 때 자료의 위치와 함께 제공자의 status와 capability를 monitoring 함으로써 실제 제공할 수 있는 또는 제공하고자 하는 속도와 다운받을 서버와 가져올 데이터의 용량을 결정함으로써 대용량 데이터를 단시간에 효율적으로 주고받을 수 있다.

• Journal of Advanced Navigation Technology
• /
• v.15 no.4
• /
• pp.632-639
• /
• 2011

Searching distributed resource efficiently is very important in distributed computing, cloud computing environment. Distributed resource searching may have system overheads and take much time in proportion to the searching number, because distributed resource searching has to circuit many peers for searching information. The open-source community project JXTA defines an open set of standard protocols for ad hoc, pervasive, peer-to-peer computing as a common platform for developing a wide variety of decentralized network applications. In this paper, we proposed peer search algorithm based on JXTA-Sim. original JXTA peer searching algorithm selected a loosely-consistent DHT. Our Lookup algorithm decreases message number of WALK_LOOKUP and reduce the network system overload, and we make a result of same performance both original algorithm and our proposed algorithm.

• Journal of KIISE:Computing Practices and Letters
• /
• v.7 no.6
• /
• pp.630-640
• /
• 2001

In this paper, we describe the design and implementation of ObjectPeerWork, which is a framework for the development of shared object model-based P2P(Peer-to-Peer) applications. The shared object model can prevent the computing power decrease on the way of resource management by incorporating the resource management function into resources themselves, and raise reliability on shared resources by improving the security problems. Also this model assures expandability by means of distributed component-based request broker manager and module container. The ObjectPeerWork based on this shared object model is a framework which makes the implementation of the enterprise information system possible, and makes distribution of the computing power and efficient resource management possible by improving the weakness in the general P2P model.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.06a
• /
• pp.481-487
• /
• 2001

본 논문은 현재까지 제시되어진 peer to peer network model들을 정리하고 대표적인 peer to peer network model에 지문인식을 적용하여 개인에 대한 신원 인증 절차를 수행함으로써 보안에 대한 새로운 해결책을 제안하였다. 기존의 peer to peer network model은 개인 대 개인간의 효율적인 network검색 기능과 분산 computing 환경을 제공하지만 보안에 관해서는 아직까지도 많은 연구가 필요하다. 본 연구에서는 기존의 peer to peer network model들에 지문인식을 사용한 새로운 보안 model을 설계하였다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2002.05a
• /
• pp.383-390
• /
• 2002

본 논문에서는 Peer-to-Peer 환경에서의 협동적 필터링의 필요성에 대해 논의하고, 이를 위한 P2P 컴퓨팅의 구조 설계, Peer Clustering기법, Peer간 메시지의 종류, 그리고 이를 구현한 시스템에 대하여 설명한다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.8
• /
• pp.586-590
• /
• 2009

The objective of this paper describes the design of MapReduce over Peer-to-Peer network for dynamic environments applications. MapReduce is a software framework used for Cloud Computing which processing large data sets in a highly-parallel way. Based on the Peer-to-Peer network character which node failures will happen anytime, we focus on using a DHT routing protocol which named Pastry to handle the problem of node failures. Our results are very promising and indicate that the framework could have a wide application in P2P network systems while maintaining good computational efficiency and scalability. We believe that, P2P networks and parallel computing emerge as very hot research and development topics in industry and academia for many years to come.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2005.05a
• /
• pp.477-482
• /
• 2005

Recently, client computing trend has been changed from server oriented information application to network based P2P(Peer to Peer) services. The conventional client/server method has the merit of accessing abundant information, on the other side P2P has the merit of synchronized community support and information exchange. P2P has four meaning of point to point, peer to peer, person to person and path to profitability. In manufacturing system field, the second meaning is interested. P2P is classified to three type such as conventional client/server, hybrid P2P and pure P2P. The third is really peer to peer concept. The related technologies with P2P are P2P searching, XML, cooperation, IPv6, computing sharing and P2P communication. This paper describes the scheme of P2P and related contents. And through the P2P based technology, a P2P-based collaborative machine and a vertical portal machine are introduced in this paper. The scheme of the machines mentioned above is suggested for cooperation in manufacturing system and u-Manufacturing.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.28-30
• /
• 2001

클라이언트-서버 환경에서 공간 데이터와 같은 대용량의 데이터를 처리하는 시스템이나 다수의 클라이언트의 요구가 발생하는 시스템에서는 데이터 처리 시 발생하는 서버 I/O 연산의 수행 비용과 질의 처리비용 및 결과 데이터의 전송 비용이 서버 사이드의 병목 현상과 질의 처리속도의 저하라는 문제점을 유발한다. 본 논문에서는 이러한 문제점을 개선하기 위한 방법으로 서버 처리비용의 분산을 위한 확장된 Peer-to-peer를 지원하는 방식을 제안하며, 이러한 확장 된 Peer-to-peer방식을 지원할 수 있는 공간데이터 관리기의 설계 및 구현상황을 제안한다. 공간데이터 관리기는 서버에 접속된 클라이언트의 정보와 클라이언트에 캐쉬된 데이터의 정보를 관리하는 공간데이터 관리기의 CIT(Client Information Table)와 이 CIT에 가용한 데이트를 캐쉬한 클라이언트가 여럿인 경웨 대상 클라이언트의 선정이 요구되며, 적은 비용으로 처리 가능한 클라인트의 선정의 위한 알고리즘과 클라이언트 사이의 확장된 Peer-to-peer 방식을 지원하기 위한 서버와 클라이언트간의 데이터 일관성 유지를 위한 데이터 유효성 관리를 제안한다.

• Journal of KIISE:Information Networking
• /
• v.32 no.1
• /
• pp.89-99
• /
• 2005

Advances of Internet and rC accelerate distribution and sharing of information, which make P2P(Peer-to-Peer) computing paradigm appear P2P computing Paradigm is the computing paradigm that shares computing resources and services between users directly. A fundamental problem that confronts Peer-to-Peer applications is the efficient location of the node that stoles a desired item. P2P systems treat the majority of their components as equivalent. This purist philosophy is useful from an academic standpoint, since it simplifies algorithmic analysis. In reality, however, some peers are more equal than others. We propose the P2P protocol considering differences of capabilities of computers, which is ignored in previous researches. And we examine the possibility and applications of the protocol. Simulating the Magic Square, we estimate the performances of the protocol with the number of hop and network round time. Finally, we analyze the performance of the protocol with the numerical formula. We call our p2p protocol the Magic Square. Although the numbers that magic square contains have no meaning, the sum of the numbers in magic square is same in each row, column, and main diagonal. The design goals of our p2p protocol are similar query response time and query path length between request peer and response peer, although the network information stored in each peer is not important.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.69-75
• /
• 2006

Internet-based distributed computing environment have been developed for advanced science and engineering by sharing large-scale resources. Therefore efficient scheduling algorithms for allocating user job to resources in the Internet-based distributed computing environment are required. Many scheduling algorithms have been proposed. but these algorithms are not suitable for the Internet-based Distributed computing environment. That is the previous scheduling algorithm does not consider peer self-control. In this paper, we propose a Peer Availability Period Prediction Strategy for Internet-based distributed computing environment and show that our Strategy has better performance than other Strategy through extensive simulation.