Modifying the Overlay Network of Freenet-style Peer-to-Peer Systems after Successful Request Queries

Size: px
Start display at page:

Download "Modifying the Overlay Network of Freenet-style Peer-to-Peer Systems after Successful Request Queries"

Transcription

1 Modifying the Overlay Network of Freenet-style Peer-to-Peer Systems after Successful Request Queries Jens Mache, David Ely, Melanie Gilbert, Jason Gimba, Thierry Lopez and Matthew Wilkinson Lewis & Clark College Portland, OR 97219, USA {jmache, dely, mgilbert, gimba, lopez, Abstract In most peer-to-peer systems, edge resources self-organize into overlay networks. At the core of Freenet-style peer-to-peer systems are insert and request algorithms that dynamically change the overlay network and replicate files on demand. We run simulations to test how effective these algorithms are at improving the performance of subsequent queries. Motivated by the observation that the performance of the original Freenet algorithms improves less rapidly with a (more realistic) ratio of 99 requests to 1 insert than with an equal number of requests and insert, we design and test a wide range of new request algorithms that pass word of success to neighboring nodes and add additional edges to the overlay network. Our results show that performance is sensitive to (1) where to add edges and (2) how many edges to add. In comparison to the original request algorithm, our best new algorithm is able to reduce average request pathlength by up to a factor of Introduction As the Internet continues to experience rapid growth and ever increasing numbers of people request particular pieces of information, it becomes exceedingly difficult to fulfill very popular requests (cf. Slashdot effect). Yet, to fulfill all the requests for one piece of information, there is an alternative to one powerful server (or server cluster): to rely on copies of the piece of information distributed across the network, so that many computers can fulfill these requests. This scalable and decentralized approach is an example of peer-to-peer networking. In most peer-to-peer systems, edge resources self-organize into overlay networks. We are studying the performance of Freenet-style peerto-peer systems in which every participating computer (node) automatically helps to fulfill queries by forwarding queries intelligently to one other node. At the core of Freenet-style peer-to-peer systems are insert and request algorithms that dynamically change the overlay network and replicate files on demand. We ran simulations to test how effective various request algorithms are in improving the performance of subsequent queries. Motivated by the observation that the performance of the original Freenet algorithms improves less rapidly with a (more realistic) ratio of 99 requests to 1 insert than with an equal number of requests and insert, we design and test a wide range of new request algorithms that further reward the fulfillers by passing word of success to neighboring nodes and adding additional edges to the overlay network. Our results are summarized below. It is better to spread the news deep rather than broad. It is better to start spreading the news at the requester (towards other regions of the network) rather than at the fulfiller. It is better to spread the news to a random neighbor rather than to the closest neighbor (the node in the reference table with the key closest to the key of the requested document). Farthest is worst. Adding too many additional references can backfire, causing the reference table to become saturated with recently requested files due to its finite storage size. The remainder of this paper is organized as follows: In Section 2, we provide background information and survey related work. In Section 3, we describe our experimental method. The original Freenet algorithms and their performance are discussed in Section 4. We present and test a wide range of new request algorithms that add additional edges to the overlay network after successful requests in Section 5. After a discussion in Section 6, we conclude in Section /4 $17. (C) 24 IEEE 1

2 A B C D References E F G H I B Documents References J 52 - K L M C Documents References 66 - N O P Q R 93 - D Documents References S T 88 - U 94 - V 97 - W 99 - X Documents Figure 1. How a Freenet-style peer-to-peer system forwards a request (originating at node A) for the document with key Background and related work 2.1. Freenet-style peer-to-peer systems In Freenet-style systems [3, 4], every node helps to fulfill requests by forwarding queries (that it cannot fulfill itself) to the node in its reference table with the key closest to the key of the requested document. For a small example, let s consider the reference tables of Fig. 1 and a request at node A for the document with key 91. Node A forwards the request to node B because 522 is closer to 91 than 111, 115, 118, 129 or 363 are. Node B forwards to node C. Node C forwards to node D that has document 91 stored. In order to maintain the privacy of the requester, the document is sent back the same way the request was forwarded. Moreover, the overlay network self-organizes in the following way: the intermediary nodes (C and B) and the requester (node A) add the entry 91 D to their reference table and a copy of document 91 to their datastore. Essential to good performance in a Freenet-style network is specialization, meaning that a node specializes in data within a dynamic and non-deterministic range. The network s insert and request algorithms are designed so that nodes specialize. However, specialization by itself is not enough. A node must also know about data that is different from its own area of concentration thereby reducing the pathlength of request queries. These shortcut or bridge connections allow requests to traverse areas of similar concentrations. Freenet s insert and request algorithms create shortcuts dynamically. The occurrence of both shortcuts and specialization in a network results in short pathlengths, a property of smallworld networks [16]. This term was coined when describing the social phenomena that most people are linked by short chains of acquaintances (popularly known as six degrees of separation ) Related work Freenet is very different from Napster [11] and Gnutella [5]. Napster s central broker was a single point of failure and could easily be attacked or shut down. Whereas Freenet operates depth-first, Gnutella operates breadth-first; for a Gnutella request, all neighboring nodes are contacted, which may in turn contact all their neighboring nodes. This can cause thousands of messages per request, making scalability a concern. (Kazaa [7] alleviates this problem with more powerful computers becoming search hubs.) Other peer-to-peer systems that operate depth-first are Chord [15], CAN [12], Pastry [14] and Tapestry/Oceanstore [18, 8, 13]. However, these systems can be viewed as providing a distributed hash table, where nodes have fixed identities and data is placed deterministically. As a consequence, items can be located within a bounded number of routing hops. On the other hand, securing against attack, load balancing and exploiting proximity are more difficult. Whereas modifications to the Gnutella and Pastry algorithms are described in [9] and [2], related work on modified algorithms for Freenet-style systems seems to be rare. Zhang et al. [17] tested different datastore replacement schemes, but did not modify the request algorithm. 3. Experimental method In this paper, our main concern is performance. Freenetstyle peer-to-peer systems dynamically change the overlay network and replicate files on demand. We ran simulations to test how effective the request and insert algorithms are in improving the performance of subsequent queries (actions). Our measure of merit is pathlength. Pathlength is the number of hops from requester to fulfiller. Short pathlength is important for (1) less overall bandwidth consumption, and (2) reduced probability of bad performance due to slow or unreliable links /4 $17. (C) 24 IEEE 2

3 For our experiments, we use the Aurora simulator [1]. This simulator was designed to model the specifications of Freenet as described in [3]. In the results reported, we simulate a network of 1, nodes that are initially connected in a very unfavorable, regular topology with four neighbors per node. The network is evolved over a specified number of actions. are either requests or inserts. They are interspersed randomly, given a request-to-insert ratio. Random keys are inserted at random nodes. Randomly-chosen existing keys are requested at random nodes. After every actions, the performance of the overlay network is measured using a set of special probe requests (that do not modify the network). Unless otherwise specified, each node has a datastore with a capacity of 5 documents and a routing table with up to 2 references. Hops-to-live (HTL) is 2 for inserts and 2 for requests (unless they are probe requests). Results are averaged over 1 simulation runs. The same 1 seeds are used to test different algorithms. We modified the Aurora simulator [1, 6] in the following way: First, for requests, keys are drawn from an exponential distribution. That way, some documents are far more popular than others. Second, each node updates the timestamps of the references that it is using. That way, references are replaced in least recently used fashion. 4. Original algorithms and their performance The original request algorithm for Freenet can be described by the following pseudocode. request(key, Hops_to_Live, Passer) if (a document in my store matches Key) pass Document back to Passer return SUCCESS endif if (Hops_to_Live equals ) return EXPIRED endif decrement Hops_to_Live while(references left in my routing table) in routing table find node with closest key that has not previously been tried answer=request(key, &Hops_to_Live, me) if(answer==expired) return EXPIRED endif if(answer==success) add Document to my store //caching add reference pointing to Fulfiller pass Document back to Passer return SUCCESS endif endwhile return BACKTRACK endrequest The insert algorithm is very similar: each node passes the document on to the node in its routing table whose key is closest to the key of the document being inserted. Then, new references are added to the routing tables. While requests terminate upon finding a copy of the document, inserts typically fulfill their hops-to-live. An insert will only stop early when attempting to insert data to a node that already has it. This is called a collision. As more requests and inserts are processed, new references are added and the routing within the network becomes more efficient. For the example of Fig. 1, nodes A, B and C added the reference 91 D. A subsequent request for 9 would find a one-hop route to D from A, B and C. As reported in [6], over time the median (and average) pathlength improves. However, the experiments in [6] assume an equal number of inserts to requests in the network. This scenario is not very realistic: in most situations (including the Internet in general), far more users request files than contribute files. Our results show that pathlength improves much slower with a more realistic ratio of 99 requests to 1 insert than with an equal number of requests and inserts (see Fig. 2). Looking closer at the simulation statistics, we notice that (1) many requests failed, and (2) successful requests added only a few new references. For simulations involving 5% requests and 5% inserts (5:5) at 5, actions, 6.9% of request-queries failed. On average, successful request-queries are adding only about 4.5 new references. This is in comparison to 2 new references added by a non-collision insert. These observations were our motivation to design and test new request algorithms. In [1], our main contribution was the design and test of strategies to improve failed (expired) request-queries by taking advantage of the work already done by the standard request process. In this paper, we focus on improving successful requestqueries by adding additional edges to the overlay network. 5. New algorithms and their performance Motivated by our observation that Freenet s original request algorithm is not very effective in changing the overlay network (because on average only a small number of new references are added per successful request, see Section 4), we set out to design and test new request algorithms. Our idea is to further reward fulfillers by adding additional references about the fulfiller and the requested document, thereby spreading the good news of a successfully fulfilled request to other areas of the network. The fulfiller is rewarded because its community of similar nodes, a natural result from the specialization Freenet promotes, will gain a bridge to another community of nodes. Hence, future searches for similar files will be faster and more efficient. In the example of Fig. 1, the reference we would like to spread further is 91 D. 91 is the key of the document that was requested and D is the fulfiller. If this information is known more widely in the network, the performance of subsequent requests potentially improves. Whereas it is rather obvious what the additional references should be about (the fulfiller and the requested key), we are faced with the question of where such references should be added, and in what quantity. To answer these questions and to improve Freenet, we are designing and testing a broad range of new algorithms, fol /4 $17. (C) 24 IEEE 3

4 :1 Requests-To-Inserts 5:5 Requests-To-Inserts 35 3 Median Pathlength Figure 2. Pathlength decreases as the network gets used. However, the ratio of requests-to-inserts makes a big difference. Requester Fulfiller Requester Fulfiller Requester Fulfiller Requester Figure 3. Breadth at fulfiller (left) and breadth at requester (right). Fulfiller Figure 4. Depth at fulfiller (right) and depth at requester (left). lowing four main approaches: breadth at requester, breadth at fulfiller, depth at requester and depth at fulfiller. In each main approach, we have the subchoices of random, closest, and farthest Breadth The new request algorithms following the breadth approach add the additional references only to nodes that are one hop away, hence behaving in a breadth-first fashion. We can specify how these neighbors are chosen from the reference table, whether at random, the closest reference to the requested key or the farthest. As shown in Fig. 3, the breadth approach can either start at the fulfiller or at the requester. The breadth fulfiller algorithm announces the news breadth-first from the fulfiller, whereas breadth requester starts at the requester. Fig. 5 and Fig. 6 show our results. Within either the breadth fulfiller or the breadth requester approach, the new request al- gorithm that chooses neighbors at random performs best. The best breadth algorithms overall is breadth requester random, see Fig Depth The new request algorithms following the depth approach add the additional references in a depth-first fashion. As shown in Fig. 4, the depth approach can either start at the fulfiller or at the requester. Again, we can specify how the next node is chosen from the reference table, whether at random, the closest reference to the requested key or the farthest. Fig. 7 and Fig. 8 show our results. Within either the depth fulfiller or the depth requester approach, the new request algorithm that chooses neighbors at random performs best. Fig. 9 shows the performance of all random algorithms. Depth requester random performs best overall /4 $17. (C) 24 IEEE 4

5 6 5 Comparison of Breadth Fulfiller Experiments Breadth Fulfiller Closest Breadth Fulfiller Farthest Breadth Fulfiller Random Figure 5. Performance of three different breadth at fulfiller algorithms. 6 5 Comparison of Breadth Requester Experiments Breadth Requester Closest Breadth Requester Farthest Breadth Requester Random Figure 6. Performance of three different breadth at requester algorithms /4 $17. (C) 24 IEEE 5

6 6 5 Comparison of Depth Fulfiller Experiments Depth Fulfiller Closest Depth Fulfiller Farthest Depth Fulfiller Random Figure 7. Performance of three different depth at fulfiller algorithms. 6 5 Comparison of Depth Requester Experiments Depth Requester Closest Depth Requester Farthest Depth Requester Random Figure 8. Performance of three different depth at requester algorithms /4 $17. (C) 24 IEEE 6

7 6 5 Comparison of All Random Experiments Breadth Fulfiller Random Breadth Requester Random Depth Fulfiller Random Depth Requester Random Figure 9. Comparison of the four random algorithms. 6. Discussion 6.1. Where to add references Our results indicate that it is better to start adding additional references at the requester rather than at the fulfiller. An explanation is that the fulfiller and its neighborhood typically already know much more about the requested key than the requester and its neighborhood. It is beneficial to announce the news towards other regions of the network. Our results also indicate that the depth-first approach is better than the breadth-first approach. An explanation is that depth-first is better at building bridges between existing specialization clusters and other clusters. In other words, we are spreading the news of the fulfiller s cluster into other specialization clusters. Breadthfirst, however, only sends the good news to our immediate neighbors in our current cluster. Thus, we are not significantly decreasing pathlengths because bridges are not being constructed. Our results furthermore indicate that random is the best criterion for choosing nodes to add the additional references to. Random is a fusion between closest and farthest. Whereas closest only sends the news essentially within the specialization cluster, and farthest indiscriminately sends the news far away and fails to reinforce the clustering action, random is the best of both methods The bread crumb algorithm Based on the lessons we learned, we designed yet another request algorithm called bread crumb. Recall the story of Hänsel and Gretel. In that story, the two children leave a trail of bread crumbs behind them so that they can later find their way home by retracing the steps along the bread crumb trail. The bread crumb algorithm works in a similar fashion. As Freenet retraces its steps along the trail that a successful request query created, bread crumb spreads the news along branches off of the main trail. Besides specifying how the next node is chosen from the reference table (random, closest or farthest), the bread crumb algorithm family has variants that go broad and/ or deep at each node along the main trail, see Fig. 1. Fig. 11 shows that random performs better than closest, as before. The best performing bread crumb variant is 2-broad 2-deep. Fig. 12 compares bread crumb 2-broad 2-deep to Freenet s original algorithm and to depth requester random, the best algorithm so far. In comparison to Freenet s original request algorithm, bread crumb 2-broad 2-deep is able to reduce average request pathlength by up to a factor of After 11,7 actions, bread crumb 2-broad 2-deep average pathlength is 6.63, whereas Freenet s original algorithms is at How many references to add The variants of bread crumb (see Fig. 1) differ in how many additional references they add. Assuming subchoice closest and routing table capacity of only references, Fig. 13 shows that average pathlength for the 3-broad 2-deep algorithm increases considerably after 2, actions. In this experiment, adding only two additional references per hop performs best. Similar trends can be seen with the other algorithm families /4 $17. (C) 24 IEEE 7

8 Original Freenet 99:1 Depth Requester Random Bread Crumb (2-broad 2-deep) Figure 12. Comparison of the best bread crumb, depth requester random and Freenet s original algorithm (logarithmic y-axis). 7 6 Bread Crumb (2-broad) Random Closest Requester Fulfiller Requester Fulfiller Requester Fulfiller Requester Fulfiller Figure 1. Variants of the bread crumb family: 1-broad (upper left), 2-broad (upper right), 2-broad 2-deep (lower left) and 3-broad 2- deep (lower right) Figure 11. Performance of two different bread crumb algorithms /4 $17. (C) 24 IEEE 8

9 Bread crumb (closest) 1-broad (1 reference per hop) 2-broad (2 references per hop) 2-broad 2-deep (6 references per hop) 3-broad 2-deep (12 references per hop) Figure 13. Adding too many additional references can backfire (variants of bread crumb, routing table capacity of only references). Depth Requester Random 1 reference per hop 3 references per hop 6 references per hop 12 references per hop Figure 14. Adding too many additional references can backfire ( depth requester random ) /4 $17. (C) 24 IEEE 9

10 Fig. 14 shows depth requester random assuming routing table capacity of 2 references (our standard assumption). In this graph, we vary the number of additional references from one per hop to 12 per hop. (In all earlier graphs, three additional references per hop were added.) Whereas initially, average pathlength decreases at a slower pace if adding only one reference per hop, after 15, actions, average pathlength increases if 12 or 6 references are added per hop. Clearly, adding too many additional references can backfire. Since reference tables are finite, some references will be pushed out. By adding too many references, we risk placing too much emphasis on recently requested files. Future work includes further investigation into how the optimal number of additional references depends on the size of the reference table, the number of nodes in the network and the number and popularity of documents. 7. Conclusions At the core of Freenet-style peer-to-peer systems are insert and request algorithms that dynamically change the overlay network and replicate files on demand. We ran simulations to test how effective these algorithms are at improving the performance of subsequent queries. We observed that Freenet s original request algorithm is not very effective in changing the overlay network because in average only a small number of new references are added per successful request. We therefore designed and tested a wide range of new request algorithms that reward fulfillers of successful requests by adding additional edges to the overlay network. Our results showed that the performance is sensitive to (1) where to add edges and (2) how many edges to add. Our main conclusions are as follows: It is better to spread the news deep rather than broad. It is better to start spreading the news at the requester (towards other regions of the network) rather than at the fulfiller. It is better to spread the news to a random neighbor rather than to the closest neighbor (the node in the reference table with the key closest to the key of the requested document). Farthest is worst. Adding too many additional references can backfire, causing the reference table to become saturated with recently requested files due to its finite storage size. In comparison to the original request algorithm, our best new algorithm, bread crumb 2-broad 2-deep, is able to reduce average request pathlength by up to a factor of Future work includes the design of distributed algorithms that dynamically change the number of additional references. Acknowledgements We would like to thank the John S. Rogers Science Research Program and the W.M. Keck Foundation for their support. References [1] Aurora simulator. viewcvs.cgi/freenet/aurora/. [2] M. Castro, P. Druschel, Y. C. Hu, and A. Rowstron. Exploiting network proximity in peer-to-peer overlay networks. In Proceedings of the International Workshop on Future Directions in Distributed Computing (FuDiCo), 22. [3] I. Clarke, S. G. Miller, T. W. Hong, O. Sandberg, and B. Wiley. Protecting free expression online with Freenet. IEEE Internet Computing, pages 4 49, January-February 22. [4] I. Clarke, O. Sandberg, B. Wiley, and T. Hong. Freenet: A distributed anonymous information storage and retrieval system. In H. Federrath, editor, Designing Privacy Enhancing Technologies, volume 29 of Lecture Notes in Computer Science, pages Springer-Verlag, 21. [5] Gnutella. [6] T. Hong. Performance. In A. Oram, editor, Peer-to-Peer Harnessing the Power of Disruptive Technologies, chapter 14, pages O Reilly, 21. [7] Kazaa. [8] J. Kubiatowicz, D. Bindel, Y. Chen, S. Czerwinski, P. Eaton, D. Geels, R. Gummadi, S. Rhea, H. Weatherspoon, W. Weimer, C. Wells, and B. Zhao. Oceanstore: An architecture for global-scale persistent storage. In Proceedings of the 9th International Conference on Architectural Support for Programming Languages and Operating Systems (ASP- LOS), 2. [9] Q. Lv, P. Cao, E. Cohen, K. Li, and S. Shenker. Search and replication in unstructured peer-to-peer networks. In Proceedings of the 16th ACM International Conference on Supercomputing (ICS), 22. [1] J. Mache, M. Gilbert, J. Guchereau, J. Lesh, F. Ramli, and M. Wilkinson. Request algorithms in Freenet-style peer-topeer systems. In Proceedings of the 2nd IEEE International Conference on Peer-to-Peer Computing, 22. [11] Napster. [12] S. Ratnasamy, P. Francis, M. Handley, R. Karp, and S. Shenker. A scalable content-addressable network. In Proceedings of ACM SIGCOMM, 21. [13] S. Rhea, C. Wells, P. Eaton, D. Geels, B. Zhao, H. Weatherspoon, and J. Kubiatowicz. Maintenance-free global data storage. IEEE Internet Computing, pages 4 49, September October 21. [14] A. Rowstron and P. Druschel. Pastry: Scalable, decentralized object location and routing for large-scale peer-to-peer systems. In Proceedings of the 18th International Conference on Distributed Systems Platforms (Middleware), 21. [15] I. Stoica, R. Morris, D. Karger, M. F. Kaashoek, and H. Balakrishnan. Chord: A scalable peer-to-peer lookup service for Internet applications. In Proceedings of ACM SIGCOMM, 21. [16] D. J. Watts and S. H. Strogatz. Collective dynamics of small-world networks. Nature, 393:44 442, June [17] H. Zhang, A. Goel, and R. Govindan. Using the small-world model to improve Freenet performance. In Proceedings of IEEE Infocom, 22. [18] B. Y. Zhao, J. Kubiatowicz, and A. D. Joseph. Tapestry: An infrastructure for fault-tolerant wide-area location and routing. Technical Report UCB/CSD , University of California, Berkeley, /4 $17. (C) 24 IEEE 1

Early Measurements of a Cluster-based Architecture for P2P Systems

Early Measurements of a Cluster-based Architecture for P2P Systems Early Measurements of a Cluster-based Architecture for P2P Systems Balachander Krishnamurthy, Jia Wang, Yinglian Xie I. INTRODUCTION Peer-to-peer applications such as Napster [4], Freenet [1], and Gnutella

More information

LessLog: A Logless File Replication Algorithm for Peer-to-Peer Distributed Systems

LessLog: A Logless File Replication Algorithm for Peer-to-Peer Distributed Systems LessLog: A Logless File Replication Algorithm for Peer-to-Peer Distributed Systems Kuang-Li Huang, Tai-Yi Huang and Jerry C. Y. Chou Department of Computer Science National Tsing Hua University Hsinchu,

More information

DYNAMIC TREE-LIKE STRUCTURES IN P2P-NETWORKS

DYNAMIC TREE-LIKE STRUCTURES IN P2P-NETWORKS DYNAMIC TREE-LIKE STRUCTURES IN P2P-NETWORKS Herwig Unger Markus Wulff Department of Computer Science University of Rostock D-1851 Rostock, Germany {hunger,mwulff}@informatik.uni-rostock.de KEYWORDS P2P,

More information

A Super-Peer Based Lookup in Structured Peer-to-Peer Systems

A Super-Peer Based Lookup in Structured Peer-to-Peer Systems A Super-Peer Based Lookup in Structured Peer-to-Peer Systems Yingwu Zhu Honghao Wang Yiming Hu ECECS Department ECECS Department ECECS Department University of Cincinnati University of Cincinnati University

More information

Shaking Service Requests in Peer-to-Peer Video Systems

Shaking Service Requests in Peer-to-Peer Video Systems Service in Peer-to-Peer Video Systems Ying Cai Ashwin Natarajan Johnny Wong Department of Computer Science Iowa State University Ames, IA 500, U. S. A. E-mail: {yingcai, ashwin, wong@cs.iastate.edu Abstract

More information

Distributed Hash Table

Distributed Hash Table Distributed Hash Table P2P Routing and Searching Algorithms Ruixuan Li College of Computer Science, HUST rxli@public.wh.hb.cn http://idc.hust.edu.cn/~rxli/ In Courtesy of Xiaodong Zhang, Ohio State Univ

More information

Subway : Peer-To-Peer Clustering of Clients for Web Proxy

Subway : Peer-To-Peer Clustering of Clients for Web Proxy Subway : Peer-To-Peer Clustering of Clients for Web Proxy Kyungbaek Kim and Daeyeon Park Department of Electrical Engineering & Computer Science, Division of Electrical Engineering, Korea Advanced Institute

More information

Building a low-latency, proximity-aware DHT-based P2P network

Building a low-latency, proximity-aware DHT-based P2P network Building a low-latency, proximity-aware DHT-based P2P network Ngoc Ben DANG, Son Tung VU, Hoai Son NGUYEN Department of Computer network College of Technology, Vietnam National University, Hanoi 144 Xuan

More information

A Framework for Peer-To-Peer Lookup Services based on k-ary search

A Framework for Peer-To-Peer Lookup Services based on k-ary search A Framework for Peer-To-Peer Lookup Services based on k-ary search Sameh El-Ansary Swedish Institute of Computer Science Kista, Sweden Luc Onana Alima Department of Microelectronics and Information Technology

More information

Dynamic Load Sharing in Peer-to-Peer Systems: When some Peers are more Equal than Others

Dynamic Load Sharing in Peer-to-Peer Systems: When some Peers are more Equal than Others Dynamic Load Sharing in Peer-to-Peer Systems: When some Peers are more Equal than Others Sabina Serbu, Silvia Bianchi, Peter Kropf and Pascal Felber Computer Science Department, University of Neuchâtel

More information

Effect of Links on DHT Routing Algorithms 1

Effect of Links on DHT Routing Algorithms 1 Effect of Links on DHT Routing Algorithms 1 Futai Zou, Liang Zhang, Yin Li, Fanyuan Ma Department of Computer Science and Engineering Shanghai Jiao Tong University, 200030 Shanghai, China zoufutai@cs.sjtu.edu.cn

More information

Should we build Gnutella on a structured overlay? We believe

Should we build Gnutella on a structured overlay? We believe Should we build on a structured overlay? Miguel Castro, Manuel Costa and Antony Rowstron Microsoft Research, Cambridge, CB3 FB, UK Abstract There has been much interest in both unstructured and structured

More information

Update Propagation Through Replica Chain in Decentralized and Unstructured P2P Systems

Update Propagation Through Replica Chain in Decentralized and Unstructured P2P Systems Update Propagation Through Replica Chain in Decentralized and Unstructured PP Systems Zhijun Wang, Sajal K. Das, Mohan Kumar and Huaping Shen Center for Research in Wireless Mobility and Networking (CReWMaN)

More information

An Agenda for Robust Peer-to-Peer Storage

An Agenda for Robust Peer-to-Peer Storage An Agenda for Robust Peer-to-Peer Storage Rodrigo Rodrigues Massachusetts Institute of Technology rodrigo@lcs.mit.edu Abstract Robust, large-scale storage is one of the main applications of DHTs and a

More information

Evolution of Peer-to-peer algorithms: Past, present and future.

Evolution of Peer-to-peer algorithms: Past, present and future. Evolution of Peer-to-peer algorithms: Past, present and future. Alexei Semenov Helsinki University of Technology alexei.semenov@hut.fi Abstract Today peer-to-peer applications are widely used for different

More information

A Chord-Based Novel Mobile Peer-to-Peer File Sharing Protocol

A Chord-Based Novel Mobile Peer-to-Peer File Sharing Protocol A Chord-Based Novel Mobile Peer-to-Peer File Sharing Protocol Min Li 1, Enhong Chen 1, and Phillip C-y Sheu 2 1 Department of Computer Science and Technology, University of Science and Technology of China,

More information

BOOTSTRAPPING LOCALITY-AWARE P2P NETWORKS

BOOTSTRAPPING LOCALITY-AWARE P2P NETWORKS BOOTSTRAPPING LOCALITY-AWARE PP NETWORKS Curt Cramer, Kendy Kutzner, and Thomas Fuhrmann Institut für Telematik, Universität Karlsruhe (TH), Germany {curt.cramer kendy.kutzner thomas.fuhrmann}@ira.uka.de

More information

A Directed-multicast Routing Approach with Path Replication in Content Addressable Network

A Directed-multicast Routing Approach with Path Replication in Content Addressable Network 2010 Second International Conference on Communication Software and Networks A Directed-multicast Routing Approach with Path Replication in Content Addressable Network Wenbo Shen, Weizhe Zhang, Hongli Zhang,

More information

Exploiting Semantic Clustering in the edonkey P2P Network

Exploiting Semantic Clustering in the edonkey P2P Network Exploiting Semantic Clustering in the edonkey P2P Network S. Handurukande, A.-M. Kermarrec, F. Le Fessant & L. Massoulié Distributed Programming Laboratory, EPFL, Switzerland INRIA, Rennes, France INRIA-Futurs

More information

Athens University of Economics and Business. Dept. of Informatics

Athens University of Economics and Business. Dept. of Informatics Athens University of Economics and Business Athens University of Economics and Business Dept. of Informatics B.Sc. Thesis Project report: Implementation of the PASTRY Distributed Hash Table lookup service

More information

Query Processing Over Peer-To-Peer Data Sharing Systems

Query Processing Over Peer-To-Peer Data Sharing Systems Query Processing Over Peer-To-Peer Data Sharing Systems O. D. Şahin A. Gupta D. Agrawal A. El Abbadi Department of Computer Science University of California at Santa Barbara odsahin, abhishek, agrawal,

More information

SCAR - Scattering, Concealing and Recovering data within a DHT

SCAR - Scattering, Concealing and Recovering data within a DHT 41st Annual Simulation Symposium SCAR - Scattering, Concealing and Recovering data within a DHT Bryan N. Mills and Taieb F. Znati, Department of Computer Science, Telecommunications Program University

More information

Simulations of Chord and Freenet Peer-to-Peer Networking Protocols Mid-Term Report

Simulations of Chord and Freenet Peer-to-Peer Networking Protocols Mid-Term Report Simulations of Chord and Freenet Peer-to-Peer Networking Protocols Mid-Term Report Computer Communications and Networking (SC 546) Professor D. Starobinksi Brian Mitchell U09-62-9095 James Nunan U38-03-0277

More information

Scalability In Peer-to-Peer Systems. Presented by Stavros Nikolaou

Scalability In Peer-to-Peer Systems. Presented by Stavros Nikolaou Scalability In Peer-to-Peer Systems Presented by Stavros Nikolaou Background on Peer-to-Peer Systems Definition: Distributed systems/applications featuring: No centralized control, no hierarchical organization

More information

A Hybrid Peer-to-Peer Architecture for Global Geospatial Web Service Discovery

A Hybrid Peer-to-Peer Architecture for Global Geospatial Web Service Discovery A Hybrid Peer-to-Peer Architecture for Global Geospatial Web Service Discovery Shawn Chen 1, Steve Liang 2 1 Geomatics, University of Calgary, hschen@ucalgary.ca 2 Geomatics, University of Calgary, steve.liang@ucalgary.ca

More information

AOTO: Adaptive Overlay Topology Optimization in Unstructured P2P Systems

AOTO: Adaptive Overlay Topology Optimization in Unstructured P2P Systems AOTO: Adaptive Overlay Topology Optimization in Unstructured P2P Systems Yunhao Liu, Zhenyun Zhuang, Li Xiao Department of Computer Science and Engineering Michigan State University East Lansing, MI 48824

More information

Architectures for Distributed Systems

Architectures for Distributed Systems Distributed Systems and Middleware 2013 2: Architectures Architectures for Distributed Systems Components A distributed system consists of components Each component has well-defined interface, can be replaced

More information

UC Berkeley UC Berkeley Previously Published Works

UC Berkeley UC Berkeley Previously Published Works UC Berkeley UC Berkeley Previously Published Works Title Impact of neighbor selection on performance and resilience of structured P2P networks Permalink https://escholarship.org/uc/item/9tq2wn2 Authors

More information

Survey of DHT Evaluation Methods

Survey of DHT Evaluation Methods Survey of DHT Evaluation Methods Markus Meriläinen Helsinki University of Technology Markus.Merilainen@tkk.fi Abstract In this paper, we present an overview of factors affecting the performance of the

More information

Flexible Information Discovery in Decentralized Distributed Systems

Flexible Information Discovery in Decentralized Distributed Systems Flexible Information Discovery in Decentralized Distributed Systems Cristina Schmidt and Manish Parashar The Applied Software Systems Laboratory Department of Electrical and Computer Engineering, Rutgers

More information

A Search Theoretical Approach to P2P Networks: Analysis of Learning

A Search Theoretical Approach to P2P Networks: Analysis of Learning A Search Theoretical Approach to P2P Networks: Analysis of Learning Nazif Cihan Taş Dept. of Computer Science University of Maryland College Park, MD 2742 Email: ctas@cs.umd.edu Bedri Kâmil Onur Taş Dept.

More information

Peer-to-Peer Systems. Chapter General Characteristics

Peer-to-Peer Systems. Chapter General Characteristics Chapter 2 Peer-to-Peer Systems Abstract In this chapter, a basic overview is given of P2P systems, architectures, and search strategies in P2P systems. More specific concepts that are outlined include

More information

Effective File Replication and Consistency Maintenance Mechanism in P2P Systems

Effective File Replication and Consistency Maintenance Mechanism in P2P Systems Global Journal of Computer Science and Technology Volume 11 Issue 16 Version 1.0 Type: Double Blind Peer Reviewed International Research Journal Publisher: Global Journals Inc. (USA) Online ISSN: 0975-4172

More information

Improving Data Availability through Dynamic Model-Driven Replication in Large Peer-to-Peer Communities

Improving Data Availability through Dynamic Model-Driven Replication in Large Peer-to-Peer Communities Improving Data Availability through Dynamic Model-Driven Replication in Large Peer-to-Peer Communities Kavitha Ranganathan, Adriana Iamnitchi, Ian Foster Department of Computer Science, The University

More information

Neighborhood Signatures for Searching P2P Networks

Neighborhood Signatures for Searching P2P Networks Neighborhood Signatures for Searching P2P Networks Mei Li Wang-Chien Lee Anand Sivasubramaniam Department of Computer Science and Engineering Pennsylvania State University University Park, PA 16802 E-Mail:

More information

A New Adaptive, Semantically Clustered Peer-to-Peer Network Architecture

A New Adaptive, Semantically Clustered Peer-to-Peer Network Architecture A New Adaptive, Semantically Clustered Peer-to-Peer Network Architecture 1 S. Das 2 A. Thakur 3 T. Bose and 4 N.Chaki 1 Department of Computer Sc. & Engg, University of Calcutta, India, soumava@acm.org

More information

Small World Overlay P2P Networks

Small World Overlay P2P Networks Small World Overlay P2P Networks Ken Y. K. Hui and John C. S. Lui Department of Computer Science & Engineering The Chinese University of Hong Kong Shatin, Hong Kong ykhui,cslui @cse.cuhk.edu.hk David K.

More information

Understanding Chord Performance

Understanding Chord Performance CS68 Course Project Understanding Chord Performance and Topology-aware Overlay Construction for Chord Li Zhuang(zl@cs), Feng Zhou(zf@cs) Abstract We studied performance of the Chord scalable lookup system

More information

Location Efficient Proximity and Interest Clustered P2p File Sharing System

Location Efficient Proximity and Interest Clustered P2p File Sharing System Location Efficient Proximity and Interest Clustered P2p File Sharing System B.Ajay Kumar M.Tech, Dept of Computer Science & Engineering, Usharama College of Engineering & Technology, A.P, India. Abstract:

More information

Adaptive Replication and Replacement in P2P Caching

Adaptive Replication and Replacement in P2P Caching Adaptive Replication and Replacement in P2P Caching Jussi Kangasharju Keith W. Ross Abstract Caching large audio and video files in a community of peers is a compelling application for P2P. Assuming an

More information

Design of PriServ, A Privacy Service for DHTs

Design of PriServ, A Privacy Service for DHTs Design of PriServ, A Privacy Service for DHTs Mohamed Jawad LINA Patricia Serrano Alvarado LINA Patrick Valduriez INRIA and LINA University of Nantes {Mohamed.Jawad,Patricia.Serrano-Alvarado}@univ-nantes.fr

More information

A Structured Overlay for Non-uniform Node Identifier Distribution Based on Flexible Routing Tables

A Structured Overlay for Non-uniform Node Identifier Distribution Based on Flexible Routing Tables A Structured Overlay for Non-uniform Node Identifier Distribution Based on Flexible Routing Tables Takehiro Miyao, Hiroya Nagao, Kazuyuki Shudo Tokyo Institute of Technology 2-12-1 Ookayama, Meguro-ku,

More information

Advanced Peer-to-Peer Networking: The P-Grid System and its Applications *

Advanced Peer-to-Peer Networking: The P-Grid System and its Applications * Advanced Peer-to-Peer Networking: The P-Grid System and its Applications * Karl Aberer, Philippe Cudré-Mauroux, Anwitaman Datta, Zoran Despotovic, Manfred Hauswirth, Magdalena Punceva, Roman Schmidt, Jie

More information

Distriubted Hash Tables and Scalable Content Adressable Network (CAN)

Distriubted Hash Tables and Scalable Content Adressable Network (CAN) Distriubted Hash Tables and Scalable Content Adressable Network (CAN) Ines Abdelghani 22.09.2008 Contents 1 Introduction 2 2 Distributed Hash Tables: DHT 2 2.1 Generalities about DHTs............................

More information

Time-related replication for p2p storage system

Time-related replication for p2p storage system Seventh International Conference on Networking Time-related replication for p2p storage system Kyungbaek Kim E-mail: University of California, Irvine Computer Science-Systems 3204 Donald Bren Hall, Irvine,

More information

Data Replication under Latency Constraints Siu Kee Kate Ho

Data Replication under Latency Constraints Siu Kee Kate Ho Data Replication under Latency Constraints Siu Kee Kate Ho (siho@cs.brown.edu) Abstract To maintain good quality of service, data providers have to satisfy requests within some specified amount of time.

More information

Peer Clustering and Firework Query Model in the Peer-to-Peer Network

Peer Clustering and Firework Query Model in the Peer-to-Peer Network Peer Clustering and Firework Query Model in the Peer-to-Peer Network Cheuk Hang Ng SAR chng@cse.cuhk.edu.hk Chi Hang Chan SAR chchan@cse.cuhk.edu.hk Ka Cheung Sia SAR kcsia@cse.cuhk.edu.hk Irwin King SAR

More information

Telematics Chapter 9: Peer-to-Peer Networks

Telematics Chapter 9: Peer-to-Peer Networks Telematics Chapter 9: Peer-to-Peer Networks Beispielbild User watching video clip Server with video clips Application Layer Presentation Layer Application Layer Presentation Layer Session Layer Session

More information

Adaptive Load Balancing for DHT Lookups

Adaptive Load Balancing for DHT Lookups Adaptive Load Balancing for DHT Lookups Silvia Bianchi, Sabina Serbu, Pascal Felber and Peter Kropf University of Neuchâtel, CH-, Neuchâtel, Switzerland {silvia.bianchi, sabina.serbu, pascal.felber, peter.kropf}@unine.ch

More information

Comparing Chord, CAN, and Pastry Overlay Networks for Resistance to DoS Attacks

Comparing Chord, CAN, and Pastry Overlay Networks for Resistance to DoS Attacks Comparing Chord, CAN, and Pastry Overlay Networks for Resistance to DoS Attacks Hakem Beitollahi Hakem.Beitollahi@esat.kuleuven.be Geert Deconinck Geert.Deconinck@esat.kuleuven.be Katholieke Universiteit

More information

Thwarting Traceback Attack on Freenet

Thwarting Traceback Attack on Freenet Thwarting Traceback Attack on Freenet Guanyu Tian, Zhenhai Duan Florida State University {tian, duan}@cs.fsu.edu Todd Baumeister, Yingfei Dong University of Hawaii {baumeist, yingfei}@hawaii.edu Abstract

More information

Brocade: Landmark Routing on Overlay Networks

Brocade: Landmark Routing on Overlay Networks Abstract Brocade: Landmark Routing on Overlay Networks CS262A Fall 2001 Yitao Duan, Ling Huang University of California, Berkeley duan@cs.berkeley.edu, hlion@newton.berkeley.edu Peer-to-peer networks offer

More information

Application Layer Multicast For Efficient Peer-to-Peer Applications

Application Layer Multicast For Efficient Peer-to-Peer Applications Application Layer Multicast For Efficient Peer-to-Peer Applications Adam Wierzbicki 1 e-mail: adamw@icm.edu.pl Robert Szczepaniak 1 Marcin Buszka 1 1 Polish-Japanese Institute of Information Technology

More information

A P2P File Sharing Technique by Indexed-Priority Metric

A P2P File Sharing Technique by Indexed-Priority Metric A P2P File Sharing Technique by Indexed-Priority Metric Toshinori Takabatake and Yoshikazu Komano Abstract Recently, the improvements in processing performance of a computer and in high speed communication

More information

A Peer-to-peer Framework for Caching Range Queries

A Peer-to-peer Framework for Caching Range Queries A Peer-to-peer Framework for Caching Range Queries O. D. Şahin A. Gupta D. Agrawal A. El Abbadi Department of Computer Science University of California Santa Barbara, CA 9316, USA {odsahin, abhishek, agrawal,

More information

PAST: A large-scale, persistent peer-to-peer storage utility

PAST: A large-scale, persistent peer-to-peer storage utility PAST: A large-scale, persistent peer-to-peer storage utility Peter Druschel Rice University, Houston, TX 77005, USA druschel@cs.rice.edu Antony Rowstron Microsoft Research, Cambridge, CB2 3NH, UK antr@microsoft.com

More information

Content Overlays. Nick Feamster CS 7260 March 12, 2007

Content Overlays. Nick Feamster CS 7260 March 12, 2007 Content Overlays Nick Feamster CS 7260 March 12, 2007 Content Overlays Distributed content storage and retrieval Two primary approaches: Structured overlay Unstructured overlay Today s paper: Chord Not

More information

Reducing Outgoing Traffic of Proxy Cache by Using Client-Cluster

Reducing Outgoing Traffic of Proxy Cache by Using Client-Cluster Reducing Outgoing Traffic of Proxy Cache by Using Client-Cluster Kyungbaek Kim and Daeyeon Park Abstract: Many web cache systems and policies concerning them have been proposed. These studies, however,

More information

Flooded Queries (Gnutella) Centralized Lookup (Napster) Routed Queries (Freenet, Chord, etc.) Overview N 2 N 1 N 3 N 4 N 8 N 9 N N 7 N 6 N 9

Flooded Queries (Gnutella) Centralized Lookup (Napster) Routed Queries (Freenet, Chord, etc.) Overview N 2 N 1 N 3 N 4 N 8 N 9 N N 7 N 6 N 9 Peer-to-Peer Networks -: Computer Networking L-: PP Typically each member stores/provides access to content Has quickly grown in popularity Bulk of traffic from/to CMU is Kazaa! Basically a replication

More information

Chapter 10: Peer-to-Peer Systems

Chapter 10: Peer-to-Peer Systems Chapter 10: Peer-to-Peer Systems From Coulouris, Dollimore and Kindberg Distributed Systems: Concepts and Design Edition 4, Addison-Wesley 2005 Introduction To enable the sharing of data and resources

More information

EARM: An Efficient and Adaptive File Replication with Consistency Maintenance in P2P Systems.

EARM: An Efficient and Adaptive File Replication with Consistency Maintenance in P2P Systems. : An Efficient and Adaptive File Replication with Consistency Maintenance in P2P Systems. 1 K.V.K.Chaitanya, 2 Smt. S.Vasundra, M,Tech., (Ph.D), 1 M.Tech (Computer Science), 2 Associate Professor, Department

More information

ReCord: A Distributed Hash Table with Recursive Structure

ReCord: A Distributed Hash Table with Recursive Structure ReCord: A Distributed Hash Table with Recursive Structure Jianyang Zeng and Wen-Jing Hsu Abstract We propose a simple distributed hash table called ReCord, which is a generalized version of Randomized-

More information

PERFORMANCE ANALYSIS OF R/KADEMLIA, PASTRY AND BAMBOO USING RECURSIVE ROUTING IN MOBILE NETWORKS

PERFORMANCE ANALYSIS OF R/KADEMLIA, PASTRY AND BAMBOO USING RECURSIVE ROUTING IN MOBILE NETWORKS International Journal of Computer Networks & Communications (IJCNC) Vol.9, No.5, September 27 PERFORMANCE ANALYSIS OF R/KADEMLIA, PASTRY AND BAMBOO USING RECURSIVE ROUTING IN MOBILE NETWORKS Farida Chowdhury

More information

Telecommunication Services Engineering Lab. Roch H. Glitho

Telecommunication Services Engineering Lab. Roch H. Glitho 1 Support Infrastructure Support infrastructure for application layer Why? Re-usability across application layer protocols Modularity (i.e. separation between application layer protocol specification /

More information

Design of a New Hierarchical Structured Peer-to-Peer Network Based On Chinese Remainder Theorem

Design of a New Hierarchical Structured Peer-to-Peer Network Based On Chinese Remainder Theorem Design of a New Hierarchical Structured Peer-to-Peer Network Based On Chinese Remainder Theorem Bidyut Gupta, Nick Rahimi, Henry Hexmoor, and Koushik Maddali Department of Computer Science Southern Illinois

More information

Relaxing Routing Table to Alleviate Dynamism in P2P Systems

Relaxing Routing Table to Alleviate Dynamism in P2P Systems Relaxing Routing Table to Alleviate Dynamism in P2P Systems Hui FANG 1, Wen Jing HSU 2, and Larry RUDOLPH 3 1 Singapore-MIT Alliance, National University of Singapore 2 Nanyang Technological University,

More information

Evaluating Unstructured Peer-to-Peer Lookup Overlays

Evaluating Unstructured Peer-to-Peer Lookup Overlays Evaluating Unstructured Peer-to-Peer Lookup Overlays Idit Keidar EE Department, Technion Roie Melamed CS Department, Technion ABSTRACT Unstructured peer-to-peer lookup systems incur small constant overhead

More information

EAD: An Efficient and Adaptive Decentralized File Replication Algorithm in P2P File Sharing Systems

EAD: An Efficient and Adaptive Decentralized File Replication Algorithm in P2P File Sharing Systems EAD: An Efficient and Adaptive Decentralized File Replication Algorithm in P2P File Sharing Systems Haiying Shen Department of Computer Science and Computer Engineering University of Arkansas, Fayetteville,

More information

FUtella Analysis and Implementation of a Content- Based Peer-to-Peer Network

FUtella Analysis and Implementation of a Content- Based Peer-to-Peer Network 8th Netties Conference Technische Universität Ilmenau September 30th to October 2nd 2002 T. Zahn / H. Ritter / J. Schiller / H. Schweppe FUtella Analysis and Implementation of a Content- Based Peer-to-Peer

More information

On Name Resolution in Peer-to-Peer Networks

On Name Resolution in Peer-to-Peer Networks On Name Resolution in Peer-to-Peer Networks Xiaozhou Li C. Greg Plaxton ABSTRACT An efficient name resolution scheme is the cornerstone of any peerto-peer network. The name resolution scheme proposed by

More information

Peer Clustering and Firework Query Model

Peer Clustering and Firework Query Model Peer Clustering and Firework Query Model Cheuk Hang Ng, Ka Cheung Sia Department of Computer Science and Engineering The Chinese University of Hong Kong Shatin, N.T., Hong Kong SAR {chng,kcsia}@cse.cuhk.edu.hk

More information

A Top Catching Scheme Consistency Controlling in Hybrid P2P Network

A Top Catching Scheme Consistency Controlling in Hybrid P2P Network A Top Catching Scheme Consistency Controlling in Hybrid P2P Network V. Asha*1, P Ramesh Babu*2 M.Tech (CSE) Student Department of CSE, Priyadarshini Institute of Technology & Science, Chintalapudi, Guntur(Dist),

More information

Improving Search in Peer-to-Peer Networks

Improving Search in Peer-to-Peer Networks Improving Search in Peer-to-Peer Networks Beverly Yang Hector Garcia-Molina fbyang, hectorg@cs.stanford.edu Computer Science Department, Stanford University Abstract Peer-to-peer systems have emerged as

More information

IN recent years, the amount of traffic has rapidly increased

IN recent years, the amount of traffic has rapidly increased , March 15-17, 2017, Hong Kong Content Download Method with Distributed Cache Management Masamitsu Iio, Kouji Hirata, and Miki Yamamoto Abstract This paper proposes a content download method with distributed

More information

Security Considerations for Peer-to-Peer Distributed Hash Tables

Security Considerations for Peer-to-Peer Distributed Hash Tables Security Considerations for Peer-to-Peer Distributed Hash Tables Emil Sit and Robert Morris Laboratory for Computer Science, MIT 200 Technology Square, Cambridge, MA 02139, USA {sit,rtm}@lcs.mit.edu Abstract.

More information

Exploiting the Synergy between Peer-to-Peer and Mobile Ad Hoc Networks

Exploiting the Synergy between Peer-to-Peer and Mobile Ad Hoc Networks Exploiting the Synergy between Peer-to-Peer and Mobile Ad Hoc Networks Y. Charlie Hu, Saumitra M. Das, and Himabindu Pucha Purdue University West Lafayette, IN 47907 {ychu, smdas, hpucha}@purdue.edu Abstract

More information

An Efficient and Secure Peer-to-Peer Overlay Network

An Efficient and Secure Peer-to-Peer Overlay Network An Efficient and Secure Peer-to-Peer Overlay Network Honghao Wang, Yingwu Zhu and Yiming Hu Department of Electrical & Computer Engineering and Computer Science University of Cincinnati {wanghong, zhuy,

More information

Performance Modelling of Peer-to-Peer Routing

Performance Modelling of Peer-to-Peer Routing Performance Modelling of Peer-to-Peer Routing Idris A. Rai, Andrew Brampton, Andrew MacQuire and Laurent Mathy Computing Department, Lancaster University {rai,brampton,macquire,laurent}@comp.lancs.ac.uk

More information

Lecture-2 Content Sharing in P2P Networks Different P2P Protocols

Lecture-2 Content Sharing in P2P Networks Different P2P Protocols Lecture-2 Content Sharing in P2P Networks Different P2P Protocols Zhou Shuigeng March 10, 2005 Outline Classification of content sharing P2P systems Content sharing P2P systems Napster; Gnutella; Freenet

More information

DATA. The main challenge in P2P computing is to design and implement LOOKING UP. in P2P Systems

DATA. The main challenge in P2P computing is to design and implement LOOKING UP. in P2P Systems LOOKING UP DATA in P2P Systems By Hari Balakrishnan, M. Frans Kaashoek, David Karger, Robert Morris, and Ion Stoica The main challenge in P2P computing is to design and implement a robust and scalable

More information

Efficient Content Location in Mobile Ad hoc Networks

Efficient Content Location in Mobile Ad hoc Networks Efficient Content Location in Mobile Ad hoc Networks Jivodar B. Tchakarov Nitin H. Vaidya University of Illinois at Urbana Champaign {tchakaro,nhv}@uiuc.edu June 9, 2003 Abstract The advances in wireless

More information

Peer-to-Peer Web Caching: Hype or Reality?

Peer-to-Peer Web Caching: Hype or Reality? Peer-to-Peer Web Caching: Hype or Reality? Yonggen Mao, Zhaoming Zhu, and Weisong Shi Wayne State University {yoga,zhaoming,weisong}@wayne.edu Abstract In this paper, we systematically examine the design

More information

Implications of Neighbor Selection on DHT Overlays

Implications of Neighbor Selection on DHT Overlays Implications of Neighbor Selection on DHT Overlays Yingwu Zhu Department of CSSE, Seattle University zhuy@seattleu.edu Xiaoyu Yang Department of ECECS, University of Cincinnati yangxu@ececs.uc.edu Abstract

More information

Self-Organizing Subsets: From Each According to His Abilities, To Each According to His Needs

Self-Organizing Subsets: From Each According to His Abilities, To Each According to His Needs Self-Organizing Subsets: From Each According to His Abilities, To Each According to His Needs Amin Vahdat, Jeff Chase, Rebecca Braynard, Dejan Kostić, Patrick Reynolds, and Adolfo Rodriguez Λ Department

More information

Impact of Neighbor Selection on Performance and Resilience of Structured P2P Networks

Impact of Neighbor Selection on Performance and Resilience of Structured P2P Networks Impact of Neighbor Selection on Performance and Resilience of Structured P2P Networks Byung-Gon Chun, Ben Y. Zhao 2, and John D. Kubiatowicz Computer Science Division, U.C. Berkeley {bgchun, kubitron}@cs.berkeley.edu

More information

Proximity Based Peer-to-Peer Overlay Networks (P3ON) with Load Distribution

Proximity Based Peer-to-Peer Overlay Networks (P3ON) with Load Distribution Proximity Based Peer-to-Peer Overlay Networks (P3ON) with Load Distribution Kunwoo Park 1, Sangheon Pack 2, and Taekyoung Kwon 1 1 School of Computer Engineering, Seoul National University, Seoul, Korea

More information

A Scalable and Robust QoS Architecture for WiFi P2P Networks

A Scalable and Robust QoS Architecture for WiFi P2P Networks A Scalable and Robust QoS Architecture for WiFi P2P Networks Sathish Rajasekhar, Ibrahim Khalil, and Zahir Tari School of Computer Science and Information Technology, RMIT University, Melbourne - 3000

More information

Challenges in the Wide-area. Tapestry: Decentralized Routing and Location. Global Computation Model. Cluster-based Applications

Challenges in the Wide-area. Tapestry: Decentralized Routing and Location. Global Computation Model. Cluster-based Applications Challenges in the Wide-area Tapestry: Decentralized Routing and Location System Seminar S 0 Ben Y. Zhao CS Division, U. C. Berkeley Trends: Exponential growth in CPU, b/w, storage Network expanding in

More information

Characterizing Traffic Demand Aware Overlay Routing Network Topologies

Characterizing Traffic Demand Aware Overlay Routing Network Topologies Characterizing Traffic Demand Aware Overlay Routing Network Topologies Benjamin D. McBride Kansas State University Rathbone Hall Manhattan, KS Email: bdm@ksu.edu Caterina Scoglio Kansas State University

More information

A Peer-to-Peer Architecture to Enable Versatile Lookup System Design

A Peer-to-Peer Architecture to Enable Versatile Lookup System Design A Peer-to-Peer Architecture to Enable Versatile Lookup System Design Vivek Sawant Jasleen Kaur University of North Carolina at Chapel Hill, Chapel Hill, NC, USA vivek, jasleen @cs.unc.edu Abstract The

More information

ECS 253 / MAE 253, Lecture 8 April 21, Web search and decentralized search on small-world networks

ECS 253 / MAE 253, Lecture 8 April 21, Web search and decentralized search on small-world networks ECS 253 / MAE 253, Lecture 8 April 21, 2016 Web search and decentralized search on small-world networks Search for information Assume some resource of interest is stored at the vertices of a network: Web

More information

On Fairness, Optimal Download Performance and Proportional Replication in Peer-to-Peer Networks

On Fairness, Optimal Download Performance and Proportional Replication in Peer-to-Peer Networks On Fairness, Optimal Download Performance and Proportional Replication in Peer-to-Peer Networks Saurabh Tewari, Leonard Kleinrock Computer Science Department, University of California at Los Angeles, 4403

More information

A Server-mediated Peer-to-peer System

A Server-mediated Peer-to-peer System A Server-mediated Peer-to-peer System Kwok, S. H. California State University, Long Beach Chan, K. Y. and Cheung, Y. M. Hong Kong University of Science and Technology A peer-to-peer (P2P) system is a popular

More information

The Robustness of Content-Based Search in Hierarchical Peer to Peer Networks

The Robustness of Content-Based Search in Hierarchical Peer to Peer Networks The Robustness of Content-Based Search in Hierarchical Peer to Peer Networks ABSTRACT M. Elena Renda I.S.T.I. C.N.R. and Scuola Superiore Sant Anna I-56100 Pisa, Italy elena.renda@isti.cnr.it Hierarchical

More information

A Scalable Content- Addressable Network

A Scalable Content- Addressable Network A Scalable Content- Addressable Network In Proceedings of ACM SIGCOMM 2001 S. Ratnasamy, P. Francis, M. Handley, R. Karp, S. Shenker Presented by L.G. Alex Sung 9th March 2005 for CS856 1 Outline CAN basics

More information

Problems in Reputation based Methods in P2P Networks

Problems in Reputation based Methods in P2P Networks WDS'08 Proceedings of Contributed Papers, Part I, 235 239, 2008. ISBN 978-80-7378-065-4 MATFYZPRESS Problems in Reputation based Methods in P2P Networks M. Novotný Charles University, Faculty of Mathematics

More information

Discovery of Stable Peers in a Self-Organising Peer-to-Peer Gradient Topology

Discovery of Stable Peers in a Self-Organising Peer-to-Peer Gradient Topology Discovery of Stable Peers in a Self-Organising Peer-to-Peer Gradient Topology Jan Sacha, Jim Dowling, Raymond Cunningham, and René Meier Distributed Systems Group, Trinity College, Dublin {jsacha,jdowling,rcnnnghm,rmeier}@cs.tcd.ie

More information

A Routing Mechanism by Distance-weighted Bloom Filter *

A Routing Mechanism by Distance-weighted Bloom Filter * 7659, England, UK Journal of Information and Computing Science Vol. 2, No. 1, 2007, pp. 55-60 A Routing Mechanism by Distance-weighted Bloom Filter * Xun Duan + and Jian-shi Li School of Computer Science

More information

INF5071 Performance in distributed systems: Distribution Part III

INF5071 Performance in distributed systems: Distribution Part III INF5071 Performance in distributed systems: Distribution Part III 5 November 2010 Client-Server Traditional distributed computing Successful architecture, and will continue to be so (adding proxy servers)

More information

FPN: A Distributed Hash Table for Commercial Applications

FPN: A Distributed Hash Table for Commercial Applications FPN: A Distributed Hash Table for Commercial Applications Cezary Dubnicki, Cristian Ungureanu, Wojciech Kilian NEC Laboratories Princeton, NJ, USA {dubnicki, cristian, wkilian}@nec-labs.com Abstract Distributed

More information