Survey Systems Peer-to-peer

A systems study on peer-to-Peer

1.G.Satyavathy, Professor of Computer Science Department, Sri Ramakrishna College of Arts and Sciences for Women, Coimbatore-641 044.

2.Dr.M.Punithavalli, Director and Head of the Department of Computer Science, Sri Ramakrishna College Arts and Science for Women, Coimbatore-641 044.

ABSTRACT

In this study, we propose a framework for analyzing peer-to-peer distribution technologies content. Our approach focuses on nonfunctional characteristics such as security, scalability, performance, management of equity and resource potential and examines how these characteristics are reflected in and affected by the architecture design decisions taken by the systems peer-to-peer way. Today Peer-to-Peer (P2P) has become an important part of the Internet, millions of users have been attracted to use their facilities and the popularity of systems services.The Peer-to-Peer researchers accelerate research university to join the systems, networks and theory. The most popular P2P file sharing and distribution support content, new applications are emerging in different areas, Internet telephony is one example. This article discusses the problems of P2P systems, such as characteristics, structures, protocols, disadvantages, problems and open fields in terms of development.

Keywords: systems distributed algorithms, peer-to-peer, performance, design, grid computing, peer toPeer.

1. INTRODUCTION

  Computation in networks of processors, each with a portion of the inputs and / or resources initially calculations can be classified as centralized or distributed. A solution based on a centralized node is designated as the node of the computer treats the entire application locally. In distributed computing, the stages of application processing nodes are distributed among the participants. The aim of these systems is to minimize communication costs and computing. Distributed systems can be classified on a client-server and a P2P model. On the client-server, the server is the central recording unit, and the only provider of content and services. A customer may request a content and operation of services, without sharing any of their own services. The client-server model can be flat where all clients only communicate with a Single server or can be hierarchical for improved scalability.

Over the years, and now the client-server paradigm is the workhorse for most user applications. In recent years a new paradigm, peer-to-peer (P2P), particularly their support applications file sharing, exchange of content like music, movies and programs, but also have implemented successful distribution of information and Internet

telephony. A refined definition of Peer-to-Peer is: "A Peer-to-peer [P2P] is an organizational system the self, the autonomous peers (peers) which aims to use shared resources in a distributed environment avoiding central services network "[21]. It possible to say that the same toPeer is a fully decentralized, with self-organization use of resources. Because of the design principles fully decentralized and self-organization – as opposed to client-server paradigm – the concept of peer-to-peer "appears as design of the future. From the perspective of concepts peer-to-peer there are different challenges, for example, distributed systems, scalable and resilient new services. Statistics show that 50 percent of Internet traffic is subject to peer-to-peer in some cases up to 75 percent. Internet growth users and bandwidth demands increase a rich diversity of applications. The client-server paradigm requires a great effort and resources to meet these challenges. Internet-based applications identification three main features:  

• Scalability.
• Security and reliability.
• The flexibility and quality services.

It is difficult for client-server based applications to meet the evolution of the Internet. The centralized approach is client-server a major obstacle (bottleneck resources) is easily attacked and difficult to change because of its location on the network infrastructure. All this indicates that expresses bias is the paradigm of client-server systems to peer-toPeer systems.

Peer-to peer systems 2.UNSTRUCTURED

It was the first generation of peer-to-peer based file sharing, using an unstructured approach. Napster [11] was one of them with a strategy based on a meta and servers to search for locations of data elements, then the data is transferred directly between peers. Gnutella uses a flooding technique, a query is sent to all peers in the system until the required data is pairs. Peer-to-peer is not based on a specific service infrastructure transportation offered. On the basis of TCP or HTTP connection, shaping, recovery peer system of peer structure based on allocation and distribution of content. Standard client-server systems, content is stored and provided by a central server. Peer-to-peer networks are highly decentralized and locate a particular desired content partners and provide the IP address of peer to peer research. Download this content is run through a separate connection. In the client-server system in providing services or content server (web server time server), clients do not ask the content or service on the server. In systems peer-to-peer are all resources provided by their peers, playing the role of customers and / or server, use these results in the short term (the first syllable of the term and the second server final customer). That was in the first generation of peer peerto some who have used a centralized approach. The server is available, but unlike the client-server approach that the server stores the IP address peers, where some content is available, which reduces the load on the server (Napster [11] is an example). 0.4 Gnutella and Freenet have been decentralized, replacing the core program has been exposed. These systems are based on the desired content identifier flooding in the network, to reach a large number companions. The guys who share content to meet the demand for their peers. A major drawback is the traffic generation flood of demand. To avoid this situation, Gnutella 0.6 introduces a hierarchy of nodes called superpeers that the contents of the shop available peers connected to their addresses IP. The primary mission is to reduce these superpeers hops in the search process, reducing network traffic.

The above programs are not structured as equals, because the content stored on a node and its IP address is not related and do not follow any structure. Examples of systems unstructured peer-to-peer are Napster, Gnutella [11?], FastTrack, eDonkey, Freenet.

ITEM A-3.STRUCTURED peer systems

The challenge of development of scalable structured Peerto the attention of the community of peers-made research. Because of the advantages and opportunities of decentralized systems self-organization, the researchers focused on approaches to distribution, storage addressable content database called Distributed Hash Tables (DHT). Have been developed to provide distributed indexing, scalability, reliability and failure tolerance.Using DHT a data item can be found on the network is O (Lg n). The basis of the network and the number of peers in an approach to the structure can be developed without impact on the effectiveness of distributed applications, there is a contrast with applications described above unstructured peer-to-peer, which are typically in the best case, the complexity of linear search. There are four mechanisms most interesting and representative for the message routing and location data structured system of content distribution are: Freenet [6, 7] is a system using unstructured file number and the node to produce an estimate that a file can be located, and spread of the access channel to transmit mode requests from node to node. Chord is a system whose nodes maintain a routing table distributed in the form of a circle of identification that all nodes are allocated and associated finger table is built. CAN is a system using n-dimensional Cartesian coordinate space to implement distribution and location of the routing table, each node is responsible for an area in the reference space.Tapestry (Kadmelia pastry and [13]) are based on the data structure Plaxton mesh, which holds pointers to the network nodes whose IDs match the elements of a structure in tree identification or prefixes to a position number.

 

4. Self-organization

The term self-organization is possible to consider the autonomy, self-maintenance, optimization,

adaptation capacity, restructuring, reproduction or emergency.

4.1. Definitions

System: A system is a set of components that have relationships with each other and form a unified whole. A system is distinguished from its surroundings.

Complexity: This term is used to denote the existence of system properties that make it difficult

to describe the semantics a global behavior of systems in any language but complete

Information about components and interactions are known.

Comments: Return to the entrance of a part of the production machine, system or process (such as

produce changes in an electronic circuit to improve performance or by an automatic control that provide self-correcting measures.)

Emergency: Refers to unexpected global system properties, is not present in any individual

subsystems, that emerge from the interactions between components. [5]

Complex Systems: Complex systems are systems that interact with multiple components whose behavior can not simply infer behavior components [20].

Criticality: An assembly in which a possible chain reaction is called critical, and we say that

criticality obtained.

Hierarchy: This context is defined hierarchy as a tree with roots.

Heterarchy: A heterarchy is a type of network structure that enables a high connectivity. However, in a hierarchy each node is connected to a maximum of one parent node and zero or more

child nodes. In heterarchy, however, a node can be connected to one of the nodes surroundings.

Estigmergia: estigmergia defines a paradigm of indirect and mediated by an asynchronous communication environment.

Disruption: Disruption is a disorder that causes a compensation act, which the disorder can be experienced positively or negatively.

4.2. Characteristics of self-organization

 

Based on the above definitions, self-organizing systems can be characterized as follows:

Self-determined boundaries: The border between the system and the environment is defined by the system itself.

Independence Identity and Structure The distinction between identity and structure helps explain the flexibility and adaptability.

Interview: A self-organizing system should try to stay.

Your comments heterarchy: If a system is disturbed, try to restructure to support themselves It is also necessary that relations with its neighbors lattice.

Self-determined reaction to a disturbance selforganizing system responds when a disturbance, but we must identify and evaluate the parameters of the disturbance.

These features of the systems self-organizing P2P systems can be extended to create several basic criteria such as limits reproduction, mutability, organization, measurement and adaptability, the independence criteria and the information, or complexity reduction, random, self-organized criticality and the emergency. Adding the degree Compliance with these criteria, each system has an identity or a main feature key objective of the system. Identity P2P system was imposed from the outside (the developers) and is not self-determined.

5. Areas of application

Peer-to-peer is an alternative for handling different types of information resources, the archives of the band bandwidth, storage and CPU cycles.

5.1. Information

 

This section explains how P2P networks are deployed in the areas of information.

Presence Information: Presence information is very important in P2P applications. Provides about peers and resources are available. This is relevant to the self-organizing system. The use of the information is equally important to share the CPU cycles because for the system to know what processor is overloaded and that it is not. The pair are agents of information for other peers.

Management Document: documents are usually schemes, central organization, which allows storage of management actions and use of data. A great effort is needed to create a centralized index of relevant documents. Experience shows documents created in a company are divided between the desktop without a repository central aware of their existence. In this case, P2P networks is very useful.

Collaboration: P2P provides level of records management work group minutes.

5.2. Records

 

One feature of file sharing peer client areas is that sometimes when downloading files and servers sometimes when downloading files (seventies). A central problem in P2P systems is the search for content or files needed (research problem) [4]. Through file sharing, three models several have been developed: the demand model of flooding (Gnutella) [16, 17], the centralized directory model (Napster) and document delivery standard (Freenet) [6, 7, 14].

5.3. Bandwidth

 

Traffic networks is increasing, especially in large volumes of multimedia data, share files, so that efficient use of bandwidth subject to a significant increase. When data are centralized and a spontaneous increase in demand occurs, the bandwidth becomes the bottleneck. P2P load balancing approach increases without any administrative work additional

exploiting the channels of transmission are not fully exploited. This concept applies

in the fields of broadcasting. Sharing of bandwidth is also very well exploited split large files into smaller chunks that are downloaded by the peer application BitTorrent [8] is an application of this principle.

5.4. Storage

 

With P2P networks, storage, only a portion of PC disk space will be used. A P2P network storage is a group of computers, network-based exist, which all share the available storage in the network. Examples are the last [18], pasta [15], CFS [9], Oceanstore [12], FARSITE [1], and Intermemory [10].

5.5. CPU cycles

 

There are requirements for high performance computing in the meantime there are unused computing power, the incentive for the use of P2P applications on the computer amulet. This Thus, it is achievable low price of computing power of a supercomputer can provide. This is accomplished through the creation of a group of independent computers network, where a computer is open and all the nodes in the network are merging into a PLC.

An example [SETI @ Home 2].

6. Applications based on peer-to-peer

 

Some applications based on P2P follows:

6.1. Application-layer multicast

Early in the size of the Internet, though limited, has enabled the dissemination of a packet to each node as possible. In the Internet today, the technique of broadcasting is very expensive. It is now necessary to show a selective multicast, for example. In this area of P2P technology has helped their informal networks to achieve unlimited scalability.

6.2. GRID Computing

 

The basic objective is to support distributed computing resource sharing between individuals and institutions (OU), or entities resources within a network infrastructure. The grids are structured and the rules, but not the capacity for self-organization, fault tolerance and scalability. Moreover, P2P systems are self-organization, fault tolerance, respond very well to the lack of foot traffic from their peers, but rules. All research efforts in these areas is melting the best of both worlds. In fact, the question of how the convergence between the two concepts is still

open [3].

7. ABSTRACT: Present and Future

 

There was much work to do and there is much work to do in the area. It is possible to classify and summarize all activities and research applications, present and future.

7.1. Applications

 

7.1.1. Present

  From 2004 until today

support for various forms of communications

• Telephony.
• Streaming
• Scalable systems and flexible to name.
• Personal communications (EGE-mail).
• sharing resources between organizations.
• Context / Content aware routing.

7.1.2. The future

 

The challenges of the future applications

• Video conference.
• Distribution of learning materials.
• Services location in mobile ad hoc networks (MANET), distributed and centralized.
• Context Aware Service.
• trustworthy computing.

7.2. Disadvantages

 

Reasons against peer-to-peer.

7.2.1. Present

Until today.

• Law suits against users.
• Software patents.
• intellectual property.
• P2P requires access to fixed rates.
• End nodes are still wide Low Bandwidth.
• Digital Rights Management.
• The best effort service for most applications insufficient.

7.2.2. The future

• The lack of confidence.
• Marketing as the end of P2P.
• Integrated P2P other subjects.

7.3. Shafts

What are the current research efforts and what work research to do.

7.3.1. Today

In fact, research suggests.

semantic integration of different types of information in the database-specific data pairs.

• Quality of service criteria (consistency, availability, security, reliability).
• Legacy support overlays.
• P2P and request non-response interaction.
• DHTs highly adaptable.
• Overlay optimization.
• P2P signaling efficiency.
• The dissemination of data.
• The allocation means (mechanism and protocols) and ensure the quality of the services of P2P systems.
• Self-determination of the sources of information.
• stimulus accounting.
• P2P realistic simulator.
• reputation mechanism to decentralize.
• semantic queries.
• Efficient P2P content distribution.
• content-based retrieval queries of metadata.
• Reduction of signaling.
• data-centric algorithm for P2P.
• content management.
• Application / data integration.
• Transmission Security Trust authentication.
• stimulating the market mechanism.
• reliable messaging.
• Mobile P2P cellular / ad-hoc.

7.3.2. Challenges ahead

• Anonymous but still trading secure electronic.
• Interoperability and / vs standards.
• Real P2P information systems business.
• Real-time dissemination of P2P data.
• P2P file systems.
• Concept of trust and security dynamics.
• Set Update dynamic content.
• distributed search mechanism.
• MANET P2P technologies.
• Mobile P2P.
• Search Smart.
• Service differentiation.
• integration of P2P-GRID.

Surely there is much work and this document does not conclusions (nothing more) because everyone is just beginning. The areas of applications is enormous. There are excellent views [23, 3, 22] should be used for research and teaching.

8. REFERENCES

[1] A. Adya, WJ Bolosky, M. Castro, G. Cermak, R. Chaiken, and JR Douceur. FARSITE:

Federated, available and reliable storage environment partially reliable

2002. http://Research.microsoft.com/sn/Farsite/OSDI20002.pdf

[2] D. Anderson. SETI @ home. Chapter 5, pp 67-76. OReally 2001.

[3] S. and D. AndroutsellisTheotokis Spinellis. A survey Peer-to-peer content distribution technologies. ACM Computing Surveys, vol. 36 (4) 2004.

  [4] H. Balakrishnan, MF Kaashoek, D. Karger, R. Morris and I Stoica. Search P2P data systems. Communications of the ACM, 46 (2) 2003.

[5] JL Casti. Complexity. Encyclopaedia Britannica. 2005

[6] I. Clarke. Freenets next generation routing protocol. 2003. Http://freenet.sourceforge.net/ Index. php? page = ngrouting.

  [7] I. Clarke, SG Miller, TW Kong, O. Sandberg and B. Wiley. Protect freedom of expression online with Freenet. IEEE Internet Computing, 6 (1), pp 40-49, 2002.

[8] B. Cohen., Incentives Build Robustness in Bit-Torrent. Workshop on economic systems Peer-to-Peer, 2003.

[9] F. Dabeka, Kasshoek MF, D. Karger, R. Morris, and I. Stoica. A large storage area for cooperation with the SFC. Proceedings of the 18th ACM Symposium on Principles systems operation. pp 202-215, 2001.

[10] A. Goldberg and P. Yianilos. Intermemory Transmits files. Proceedings of the IEEE International Forum on research and developments technology in digital libraries. pp 147-156, 1998.

[11] A. Kim and L. Hoffman. Napster and other Internet-peer to-peer application. University George Washington, 2002, CITES. ist.psu.edu/kim01pricing.html.

[12] JD Krubiatowicz Bindel, Y. Chen et al. OceanStore: An architecture for scale global persistent storage. Minutes of the 9th International Conference on Architectural Support for programming languages and operating systems. 2000.

[13] p. and D. Maymounkov Mazières. Kademlia: A peer-to-Peer Information System Based on the XOR metric. International Workshop on Peer-to- Systems together. (IPTPS02), 2002.

  [14] DS Milojicic, V. Kalogerakis, R. Lukose, K. And J. Nagaraja Pruyne.-To-Peer Computing Peer. HP Technical Report HPL-2002-http: / / www.hpl.hp.com/techreports/2002/HPL-2002-57.pdf.

[15] IT Moreton Pratt, and T. Harris. Storage, mutability and the nomenclature of the pasta. 2002. http://www.cl.cam.ac.uk/users/tlh20/papers/mphpasta . pdf.

[16] M. Ripeanu .- to-peer architecture as Case Study: Gnutella network. Proceedings of the IEEE 1st Conference Computers International Peer-to-Peer, 2001.

  [17] M. And I Ripeanu Foster. Mapping the Gnutella network: properties scale peer-to-Peer Systems and implications for system design. IEEE Internet Computing, 6 (1), 2002.

  [18] A. and P. Rowstron Druschel. Storage management and caching in the past, the vast majority, persistent peer-to-Peer Storage Utility. 18th ACM SOSP01. 2001.

  [19] S. Saroiu, PK Gummadi and SD Gribble. Measure and analyze the characteristics of Napster and Gnutella hosts. Multimedia Systems, 9 (2), 2003. pp 170-184, Springer-Verlag.

[20] F. Schweitzer. Coordination of decisions in multi-agent systems in space. International Workshop And the socio-economic and physical. 2003.

  [21] R. Steinmetz and K. Wehrle .- To-Peer Networking pairs and Informatics. Informatik-Spectrum, 27 (1). Springer. 2004.

  [22] R. Steinmetz and K. Wehrle (Eds) .- footing between own systems and applications. Lecture Notes in Computer Science, LNCS 3485, Springer. 2005.

  [23] J. Van Der Merwe, D. Dawound, S. Mc Donald. A Survey of point to point among themselves for managing mobile ad hoc networks, ACM computing Survey, 39 (1), 2007

About the Author

1.G.Satyavathy,Lecturer,Department Of Computer Science,Sri Ramakrishna College Of Arts and Science For Women,Coimbatore-641 044.

2.Dr.M.Punithavalli,Director and Head,Department Of Computer Science,Sri Ramakrishna College Of Arts and Science For Women,Coimbatore-641 044.

SFO Medical Fully Automatic Digital Wrist Blood Pressure Monitor & Pulse Display with Memory and Fuzzy Logic, Free Travel Case $25.00 FDA Approved Accurate automatic blood pressure and pulse monitor with memory storage for tracking trend of blood pressure Very convenient for home monitoring and continuos monitoring Memory function to store upto 80 readings Free Travel Case Low Power with automatic power off Automatic inflation and deflation of cuff Large LCD for easy viewing with date and time ? Measures Blood Pressure and Pulse…

…

Haier HLP23E Electronic 1-1/2-Cubic-Foot Touch Pulsator Top-Loading Portable Washing Machine $257.23 Save yourself the trouble of hand washing your delicates with 4 wash cycles and 4 water level settings. Includes LED Display….