Proceedings of the 5th ACM SIGCOMM Conference on Internet Measurement - IMC '05 2005
DOI: 10.1145/1330107.1330111
|View full text |Cite
|
Sign up to set email alerts
|

Client behavior and feed characteristics of RSS, a publish-subscribe system for web micronews

Abstract: Abstract

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

3
76
0
1

Year Published

2006
2006
2013
2013

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 84 publications
(80 citation statements)
references
References 6 publications
3
76
0
1
Order By: Relevance
“…Interoperating through simple publish and subscribe invocations is especially useful for the development of large, distributed, loosely coupled systems. While there are many applications based on group communication and topic-based pub/sub protocols such as information dissemination [17,22], a large variety of emerging applications benefit from the expressiveness, filtering, distributed event correlation, and complex event processing capabilities of content-based pub/sub. These applications include RSS feed filtering [31], stock-market monitoring engines [33], system and network management and monitoring [7,20], algorithmic trading with complex event processing [10,29], business process execution [32], business activity monitoring [7] and workflow management [5].…”
Section: Introductionmentioning
confidence: 99%
“…Interoperating through simple publish and subscribe invocations is especially useful for the development of large, distributed, loosely coupled systems. While there are many applications based on group communication and topic-based pub/sub protocols such as information dissemination [17,22], a large variety of emerging applications benefit from the expressiveness, filtering, distributed event correlation, and complex event processing capabilities of content-based pub/sub. These applications include RSS feed filtering [31], stock-market monitoring engines [33], system and network management and monitoring [7,20], algorithmic trading with complex event processing [10,29], business process execution [32], business activity monitoring [7] and workflow management [5].…”
Section: Introductionmentioning
confidence: 99%
“…In practice, maintaining a separate ring per topic is very expensive, notably for nodes subscribed to many topics. However, it has been observed that subscriptions tend to be strongly correlated [15]. Our approach exploits this correlation in order to substantially lower the number of links maintained: A single link can serve as a ring link for multiple topics.…”
Section: The Dissemination Overlaymentioning
confidence: 99%
“…Many applications report benefits from using this form of interaction, such as application integration [20], financial data dissemination [2], RSS feed distribution and filtering [15], and business process management [14]. As a result, many industry standards have adopted pub/sub as part of their interfaces.…”
Section: Introductionmentioning
confidence: 99%
“…We first review the work related to the detection and use of interest correlation between users in large-scale systems. The presence of communities amongst user interests and accesses in Web search traces [5], [6], peer-to-peer file sharing systems [7] or RSS news feeds subscriptions [8] can be exhibited. The existence of a correlation of interests amongst a group of distributed users has been leveraged in a variety of contexts and for designing or enhancing various distributed systems.…”
Section: Related Workmentioning
confidence: 99%
“…Cyclon and Vicinity [28] are well known Algorithms, whereas Twinfinder is a customized version of Vicinity, we conceived and designed, where the sender transfer to each neighbor a potentially different set of profiles, in particular the ones considered the most similar to the receiver. Each protocol has been tested under several different conditions, varying the maximum number of neighbors a peer can store (in the range [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20]) and the number of nodes in the networks (during this first experimental phase in the range [1000-3000]). The results we achieved are depicted in Figure 5, this figure is in turn composed of eight subfigures organized in four rows and two columns.…”
Section: Algorithm 4 Recommendation Suggestionmentioning
confidence: 99%