A scalable content-addressable network
By Sylvia Ratnasamy, Paul Francis, Mark James Handley, Richard M. Karp, Scott Shenker. Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, 2001
7,553 citations recorded by the source corpus.
Hash tables - which map "keys" onto "values" - are an essential building block in modern software systems. We believe a similar functionality would be equally valuable to large distributed systems. In this paper, we introduce the concept of a Content-Addressable Network (CAN) as a distributed infrastructure that provides hash table-like functionality on Internet-like scales. The CAN is scalable, fault-tolerant and completely self-organizing, and we demonstrate its scalability, robustness and low-latency properties through simulation.
Subfields: Computer Networks and Communications
View Paper: https://www.semanticscholar.org/paper/680ba806b8a651e8cb2e2d64a9d6bc2325a8eea1
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Networks and Communications
- John Risson
- Tim Moors
- Ion Stoica
- Scott Shenker
- Anne-Marie Kermarrec
- Hector Garcia-Molina
- Peter Druschel
- Antony Ian Taylor Rowstron
Citations in the corpus
- Herbivore: A Scalable and Efficient Protocol for Anonymous Communication (2003, background): “Some others allow the nodes to directly pick their location [13].”
- An Infrastructure for Fault-tolerant Wide-area Location and Routing (2001, methodology): “approaches, including Pastry [10], CHORD [28], and CAN [ 22 ].”
- Replication strategies in unstructured peer-to-peer networks (2002, background): “Of these decentralized designs, some are structured in that they have close coupling between the P2P network topology and the location of data; see [12, 10, 11, 15, 2] for a sampling of these designs.”
- TAO: Facebook's Distributed Data Store for the Social Graph (2013, background): “Many contributions in distributed hash tables have focused on routing [28, 32, 25, 24].”
- A survey of autonomic communications (2006, methodology): “Overlay data structures are distributed data structures that generalize the idea of overlay networks [Ratsanamy et al. 2001, 2002; Rowstron and Druschel 2001].”
- Topologically-aware overlay construction and server selection (2002, methodology): “Several ongoing projects make use of application-level or overlay networks [1], [2], [3], [4], [5], [6], [7], [8], [9].”
- Giggle: A Framework for Constructing Scalable Replica Location Services (2002, background): “The most relevant to our current work are resource discovery systems in which resources are uniquely identified by an attribute (usually a globally unique name): CAN [35], Chord [36], Tapestry [37], Gnutella [38], Freenet [39], and Napster.”
- Network Applications of Bloom Filters: A Survey (2004, methodology): “Such an algorithm can be used in conjunction with a more expensive routing algorithm such as those suggested for speci fi c P2P networks [Druschel and Rowstron 01, Ratnasamy et al. 01, Stoica et al. 01].”
- Scribe: a large-scale and decentralized application-level multicast infrastructure (2002, background): “These results indicate that Scribe does a good job of partitioning and distributing forwarding load over all nodes: each node is responsible for forwarding multicast messages for a small number of groups, and it forwards multicast messages only to a small number of nodes.”
- PocketLens: Toward a personal recommender system (2004, methodology): “Our next architecture is based on recent advances in architectures for peer-topeer file sharing networks such as Chord, CAN, and Pastry [Stoica et al. 2001; Ratnasamy et al. 2001; Rowstron and Druschel 2001a; Druschel and Rowstron 2001].”
- Reclaiming space from duplicate files in a serverless distributed file system (2002, neutral): “Also similar is CAN [ 30 ], which uses O(d)-sized neighbor tables to route information to nodes in O(d n1/d) hops.”
- Wide-area cooperative storage with CFS (2001, background): “A number of recent peer-to-peer systems use similar approaches with similar scalability and performance, including CAN [26], PAST [28] [29], OceanStore [13] [35], and Ohaha [22].”
- Secure History Preservation Through Timeline Entanglement (2002, methodology): “We also hope to migrate away from the allto-all entanglement model, by employing recentlydevelopped, highly scalable overlay architectures such as CAN [22] and Chord [23].”
- Scaling Memcache at Facebook (2013, positive): “This paper is the latest in a series of works that have recognized the flexibility and utility of distributed keyvalue stores [1, 2, 5, 6, 12, 14, 34, 36].”
- Analysis of the evolution of peer-to-peer systems (2002, background): “These include CAN [10], Pastry [3], and Tapestry [6, 14] (based on meshes [9]).”
- Peer-to-peer support for massively multiplayer games (2004, background): “A number of peer-to-peer overlays have recently been proposed, including CAN [29], Chord [36], Tapestry [37] and Pastry [32].”
- The Peer Sampling Service: Experimental Evaluation of Unstructured Gossip-Based Implementations (2004, methodology): “A structured overlay [26, 25, 27] is by definition not dynamic so to utilize it for implementing the peer sampling service random walks or other additional techniques have to be applied.”
- Serverless Computing: One Step Forward, Two Steps Back (2018, background): “stem constructs like “threads” and “ports”: nameable endpoints in the network. Various ideas from the literature have provided aspects of this idea: actors [11], tuplespaces [4] pub/sub [18] and DHTs [21] are all examples. Chosen solutions need to incur minimal overhead on raw network performance. D”