27-06-2012, 05:27 PM
IRM: INTEGRATED FILE REPLICATION AND CONSISTENCY MAINTENANCE IN P2P SYSTEMS
IRM Doucment.doc (Size: 158 KB / Downloads: 40)
ABSTRACT:
In peer-to-peer file sharing systems, file replication and consistency maintenance are widely used techniques for high system performance. Despite significant interdependencies between them, these two issues are typically addressed separately. Most file replication methods rigidly specify replica nodes, leading to low replica utilization, unnecessary replicas and hence extra consistency maintenance overhead. Most consistency maintenance methods propagate update messages based on message spreading or a structure without considering file replication dynamism, leading to inefficient file update and hence high possibility of outdated file response. This paper presents an Integrated file Replication and consistency Maintenance mechanism (IRM) that integrates the two techniques in a systematic and harmonized manner. It achieves high efficiency in file replication and consistency maintenance at a significantly low cost. Instead of passively accepting replicas and updates, each node determines file replication and update polling by dynamically adapting to time-varying file query and update rates, which avoids unnecessary file replications and updates. Simulation results demonstrate the effectiveness of IRM in comparison with other approaches. It dramatically reduces overhead and yields significant improvements on the efficiency of both file replication and consistency maintenance approaches.
INTRODUCTION:
Over the past years, the immerse popularity of Internet has produced a significant stimulus to peer-to-peer (P2P) file sharing systems. A recent large-scale characterization of HTTP traffic has shown that more than 75 percent of Internet traffic is generated by P2P applications. The percentage of P2P traffic has increased significantly as files such as videos and audios have become almost pervasive. The study also shows that the access to these files is highly repetitive and skewed towards the most popular ones. Such objects can exhaust the capacity of anode, leading to delayed response. File replication is an effective method to deal with the problem of over load condition due to flash crowds or hot files. It distributes load over replica nodes and improves file query efficiency. File consistency maintenance to maintain the consistency between a file and its replicas is in dispensable to file replication. Requiring that the replica nodes be reliably in for me do fall updates could be prohibitively costly in a large system.
Thus, file replication should proactively reduce unnecessary replicas to minimize the over head of consistency maintenance, which in turn provides guarantee for the fidelity of consistency among file replicas considering file replication dynamism. File replication dynamism represents the condition with frequent replica node generation, deletion, and failures. Despite the significant interdependencies between file replication and consistency maintenance, they have been studied separately. In most current file replication methods, file owners rigidly specify replica nodes and the replica nodes passively accept replicas. The methods were designed without considering the efficiency of subsequent file consistency maintenance.
These methods make it difficult to adjust the number of replicas to the time-varying utilization of replicas and to ensure that all replicas are fully utilized. The number of replicas has a significant impact on the overhead of file consistency maintenance. Large number of replicas needs more updates hence high consistency maintenance overhead and vice versa. Therefore, the methods lead to high overhead for unnecessary file replications and consistency maintenance. Though these methods generally can be applied to all file replication methods, they cannot be exploited to their full potential without considering time varying and dynamic replica nodes. Structure-based methods assume relatively stable replica nodes, which does not hold true in practice due to dynamic replica nodes caused by file replication. Replicas nodes may be continuously and rapidly generated, deleted, and fail. Such file replication dynamism will lead to unsuccessful update propagation and significantly high overhead for structure maintenance. System-wide message spreading will generate tremendously unnecessary redundant messages. In addition, they cannot guarantee that all replica nodes can receive a update message. Therefore, without taking into account file replication dynamism, consistency maintenance generates unnecessary over head and cannot help to guarantee the fidelity of replica consistency. Furthermore, as in file replication, passively accepting update messages makes it difficult to avoid unnecessary updates in order to reduce overhead.
Uncoordinated deployment of file replication and consistency maintenance techniques can negate each other’s efforts and lead to suboptimal or even low system performance. As a result, on one hand, file replication is faced with a challenge to minimize the number of replicas to reduce the consistency maintenance overhead, without compromising its efficiency in hot spot and query latency reduction. On the other hand, consistency maintenance is faced with a challenge to guarantee the fidelity of replica consistency in a timely fashion with low overhead considering file replication dynamism. This makes it important to integrate the two techniques to enhance their mutual interactions and avoid their conflicting behaviors, ensuring that the two techniques can be exploited to their fullest capacities.
This paper presents an Integrated file Replication and consistency Maintenance mechanism (IRM) that achieves high efficiency in file replication and consistency maintenance at a significantly lower cost. IRM integrates file replication and consistency maintenance in a harmonized and coordinated manner. Basically, each node actively decides to create or delete a replica and to poll for update based on file query and update rates in a totally decentralized and autonomous manner. It replicates highly queried files and polls at a high frequency for frequently updated and queried files. IRM avoids unnecessary file replications and updates by dynamically adapting to time varying file query and update rates. It improves replica utilization, file query efficiency, and consistency fidelity. A significant feature of IRM is that it achieves an optimized trade-off between overhead and query efficiency as well as consistency guarantees.
EXISTING SYSTEM:
• Content distribution is a centralized one, where the content is distributed from the centralized server to all clients requesting the document.
• Clients send request to the centralized server for downloading the file.
• Server accepts the request and sends the file as response to the request.
• In most client-server setups, the server is a dedicated computer whose entire purpose is to distribute files.
PROPOSED SYSTEM:
• We proposed an efficient and adaptive centralized file replication algorithm in P2P file sharing systems called IRM. In the method, traffic hubs that carry more query load and frequently requesters are chosen as replica nodes.
• Peer-to-peer content distribution provides more resilience and higher availability through wide-scale replication of content at large numbers of peers.
• A P2P content distribution community is a collection of intermittently-connected nodes with each node contributing storage, content and bandwidth to the rest of the community
• The peer-to-peer file sharing networks had a centralized server system. This system controls traffic amongst the users.
SYSTEM SPECIFICATION:
HARDWARE SPECIFICATION:
Processor : Intel Pentium-IV
Speed : 1.1GHz
RAM : 512MB
Hard Disk : 40GB
General : Key Board, Monitor , Mouse
SOFTWARE SPECIFICATION:
Operating System : Windows XP
Software : JAVA ( JDK 1.5.0)
MODULE DESCRIPTION:
MODULE 1:
File is sharing into IRM of equal size and k simultaneous connections are used. Client downloads a file from P2P at a time. Each peer sends a replication to the client.
MODULE 2:
File is divided into many p2p and user downloads file replication sequentially one at time. The client randomly chooses the source peer at each time slot and download the file replication from each peer in the given time slots.
MODULE 3:
Whenever a user completes a replication from its current source peer, the user randomly selects a new source peer and connects to it to retrieve a new p2p. Switching source peers based on chunk can reduce average time varying file download replications and updates.
MODULE 4:
File replication is an effective method to deal with the problem of overload condition due to flash crowds or hot files. It distributes load over replica nodes and improves file query efficiency. File consistency maintenance to maintain the consistency between a file and its replicas is indispensable to file replication. Requiring that the replica nodes be reliably informed of all updates could be prohibitively costly in a large system.