Back to bookmarks

Back to index

Bookmarks: Peer-to-peer

Note: This section is much less organized than the remaining ones.

Who's who

People - Manfred Hauswirth (P-Grid, Gridella; self-organizing systems, push content delivery and event-based systems). Also produced a p2p tutorial held at several conferences, downloadable with his papers - Jon Crowcroft (grid computing and many different work on networks). Also author of a p2p tutorial held at conferences like Networking 2002 (forthcoming on Springer LNCS, downloadable from his page) - Yuan Chen (Intermemory) - Peter Yianilos (Intermemory, with papers) - Ross Anderson (Eternity, Xenoservers, Cocaine auction protocol, steganographic filesystem)

Academic and industrial research groups - Microsoft Cambridge Research on Distributed Systems; Pastry, PAST, Scribe, Farsite and links

Consortia - Grid computing forum. Absorbed the P2P Working Group of Intel et al. - Internet2 Peer-to-peer working group

Projects - PEPITO - Peer-to-Peer: Implementation and Theory, a three-year program collaboration between SICS(Sweden), EPFL (Switzerland), INRIA Rocquencourt (France), KTH (Sweden), UCL (Belgium) and the University of Cambridge (UK)

SWAP: Semantic Web And Peer-to-peer (still to approve)

Related - NAPI (Network-aware Programming and Interoperability) project home page. With link to local pages, workshops, published material - Crossflow project: Cross-organizational workflow support in virtual enterprises - DBGlobe project: A Data - centric Approach to Global Computing - Xenoservers at Cambridge

Conferences and workshops - Peer-to-peer conference 2001 - Peer-to-peer conference 2002 - 1st International Workshop on Peer-to-peer Systems (IPTPS 2002). Online proceedings - International Workshop on Peer-to-Peer Computing at Networking 2002. Online proceedings - Internet2 workshop on Collaborative Computing in Higher Education: Peer-to-Peer and Beyond. Online presentations

Related - CodeCon is the premier event in 2002 for the P2P, cypherpunk, and network/security application developer community. It is a workshop for developers of real-world applications that support individual liberties. A hacker-oriented conference with interesting things

See NAPI (above) for workshops

Resources - A generic site on p2p from O'Reilly - p2p networks status and home for reflector, a gnutellanet proxy software (defunct) - Peer-to-peer at Yahoo. Many many links - Delboy's links

See also the application list at openp2p.

Some other interesting links at the YOID project.

Applicative scenarios and applications

Sometimes cited by p2p authors as examples and fonts of inspiration

* - talk: the "ancient" chat program for Unix-like systems, here in the GNU implementation

Usenet, DNS, routing (BGP, RIP), TCP, IP

* X/Open DTP

Infrastructures, middleware, protocols... - Gnutella: more an infrastructure for search, on which a file sharing system has been built, than a file sharing system - JXTA: a set of open peer-to-peer protocols for interoperability and communication between connected devices - Fasttrack created a core peer-to-peer stack/API on which peer-to-peer applications can be built; kazaa and morpheus are based on it. Absolutely no documentation - giFT: internet File Transfer, an infrastructure for file transfer, loosely based on Fasttrack

*, - OpenCola: content management and collaborative search

* - Uprizer: content distribution; Founded by Ian Clarke, the creator of Freenet, and based on Freenet. Not much difference from Freenet

* - Adaptinet, producers of a P2P SDK

* - Affiniti produced (seemingly) a distributed p2p oo programming environment

* - Alpine: cited in Hauswirt tutorial. Alpine is Adaptive Large-scale Peer-to-peer Information NEtworking, a general purpose resource discovery infrastructure which can handle high churn rates, malicious peers, and limited bandwidth. Many resources

* - Neurogrid is a framework for finding resources in a distributed environment. Presented at Networking 2002 P2P workshop (see above)

Data management

* - HiveCache: Online backup and enterprise data storage software. Ex Mojonation (now defunct)

* - AskMe produces solution to discover, exchange, capture, dispatch and analyze Employee Knowledge Networks

File sharing - Napster (now in a redesign phase) - OpenNap: Open source Napster implementation. Useful for techincal information on Napster, - AppleSoup, FlyCode : founded by ex Napster techies, with content rights management (seemingly defunct for infeasibility due to excessive law restrictions)

* - Ohaha: file sharing. Based on Microsoft Cache Array Routing Protocol (CARP) and MIT consistent hashing ( on consistent hashing and random trees). Now retired, if you google-search you may find the old cached paper on its implementation

* Morpheus, Kazaa, Grockster: based on Fasttrack (see above)

Distributed large scale resilient data stores - Oceanstore is aimed at providing global-scale persistent data using a variant of DDLP (Distributed Data Location Protocol)

* - Archivial Intermemory. Very poor, see Yuan Chen above for a paper

* - Farsite is Federated, Available, and Reliable Storage for an Incompletely Trusted Environment

* PAST (see below, under "Pastry")

* - Eternity

with focus on security - Freenet - Free Haven by MIT: With different levels of anonimity and security - Publius - Red Rover (defunct?): not much peer-to-peer, interesting as it raises issues on security in very hostile environments. An article in O'Reilly p2p book

Farsite, PAST (see above)

* - CliqueNet is a peer-to-peer, self-organizing, scalable communication protocol that guarantees anonymity. A project at Cornell University

Messaging and collaboration

* - Jabber: messaging and notification infrastructure

AIM (AOL Instant Messenger), Aimster, ICQ

* - Groove: platform and service for intra- and inter-enterprise communication and collaboration; From the inventor of Lotus Notes

* - HIVE: collaborative workspaces based on JXTA

Business process integration systems

* - Consilient: p2p infrastructure to support creation, distribution and optimization of business processes

* - Agentware: Distributed universal transactions for business

* business process integration product suite

Realtime content distribution and streaming

* - Allcast patented a technology for p2p (scalable) streaming

Distributed grid computing - SETIAtHome

* - Avaki

* - Entropia

* - Popular Power, now out of business

* - Globus

Middleware: public-subscribe, events, multicasting - Scribe: based on Pastry (see above), system for large scale event notification

From Scribe bibliography:

* Herald: in Proceedings HotOS VIII, May 2001

* Birman et al "Bimodal multicast" ACM Transaction on Computer Systems, 17(2) 41-48 May 1999

* Eugster et al "Lightweight probabilistic broadcast" in Proceedings Conference on Dependable Systems and Networks (DNS 2001) July 2001

Works on distributed filesystems, more or less related with p2p

* - Elephant file system

* - Self certifying file system (SFS). It is also part of the MIT PDOS group, like Chord

* - Bayou: filesystem for nomadic users (disconnected operations)

Other works - Anthill: Multi-agent P2P system based on self-organization

* - Terminodes: p2p information system based on mobile terminals, for mobile adhoc networks

* - Mariposa is a distributed DBMS. It allows DBMS far apart and on different administrative domains to work together to process queries (terminated project)

* - Globe is a paradigm for building wide area distributed system based on distributed shared objects; By Tanenbaum's group (has it anything to do with p2p?)

Still unclassified

* - YOID project; YOID stands for Your Own Internet Distribution. Although on the same site of CAN seems unrelated with it. See presentation at Internet2 p2p workshop (above). Also has links at many other related projects

* - Grid ad-hoc mobile network routing protocol, from the MIT PDOS project of which also Chord is part. Do not confond with P-Grid (and with grid computing)

* - Akamai: cited in Hauswirt tutorial

* - Gnutmeg: cited in Hauswirt tutorial

* - DFSI, Distributed File System over IRC, cited in Hauswirt tutorial

* - OFSI, Open File Sharing Initiative, cited in Hauswirt tutorial (defunct?)

* India: cited in Hauswirt tutorial


Resource localization based on name (distributed hashing) - Chord is a lookup algorithm for P2P systems whose strong points are simplicity, provable correctness and provable performance

* - Circle is an *implementation* of decentralized hashtables similar to (inspired by?) Chord. Free code available - Content Addressable Network (CAN): home page of Sylvia Ratnasamy, paper in proceedings ACM SIGCOMM 2001, and more papers - P-Grid and Gridella. See also Manfred Hauswirth's page (see above under "People"). Do not confond with grid computing - Pastry from Microsoft Cambridge: also used to build PAST and Scribe, a persistent and anonymous document store, and a large scale event notification system respectively (see above)

* - Tapestry from Zhao, Kubiatowicz, Joseph (see also Berkeley technical report UCB//CSD-01-1141). A lot of links

* - Snowball and works of Radek Vingralek

* - "Search in power law networks": Efficient search in p2p networks

* Consistent hashing, random trees, CARP; see also above under "file sharing"

* See the papers of Plaxton (very cited), Litwin, Yokota

Resource discovery and search (semantically based) - Service Location Protocol homepage: Automatical discovery on IP networks (Sun e IETF). Almost concluded, soon to be abandoned (refer to IETF) - JESA service discovery protocol


* - Zero Knowledge Systems. Makers of Freedom Network

* - Cypherpunks. They produced the Cypherpunk Type 1 remailer

* - Mixmaster remailer, a Type 2 anonymous remailer

* - Orange: remailer and statistics on remailer net's health

Reputation and network economy

Golle, Leyton-Brown, Mironov "Incentives for sharing in peer-to-peer networks"

Nota: * = non ancora esaminato

Valid XHTML 1.0!

Updated: 2004-01-08 23:44 UTC