Academic
Publications
Matching events in a content-based subscription system

Matching events in a content-based subscription system,10.1145/301308.301326,Marcos Kawazoe Aguilera,Robert E. Strom,Daniel C. Sturman,Mark Astley,Tus

Matching events in a content-based subscription system   (Citations: 399)
BibTex | RIS | RefWorks Download
Content-based subscription systems are an emerging alternative totraditional publish-subscribe systems, because they permit moreflexible subscriptions along multiple dimensions. In these systems,each subscription is a predicate which may test arbitrary attributeswithin an event. However, the matching problem for content-basedsystems --- determining for each event the subset of all subscriptionswhose predicates match the event --- is still an open problem. Wepresent an efficient, scalable ...
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...While the classic \at" ad matching problem and its publishersubscriber generalization are well known and have been thoroughly studied [12, 6, 7, 16], the methods developed in that body of work do not address the crucial ingredient in the new problem, namely the requirement that the ad server be able to quickly construct optimal valid paths through an arbitrary directed graph of constraint-annotated business agreements...
    • ...[12, 6, 7, 16]) does not address the key issue in the formal ExchangeAd-Serve problem, it does contain ideas that might help to further optimize the Exchange’s real ad server...
    • ...For example, there are several plausible ways that the indexing techniques developed in [7, 6, 8] could be used to speed up the server...

    Kevin Langet al. Efficient online ad serving in a display advertising exchange

    • ...Many algorithms have been proposed to date for efficient matching of events to subscriptions in pub-sub systems [19], [20], [21], [22], [23], [24]...
    • ...An algorithm for event-matching based on the concept of subscription trees is described in context of the GRYPHON project [20]...

    Ashish Kamraet al. Design and Implementation of an Intrusion Response System for Relation...

    • ...Content-based publish/subscribe is of interest to both database and networking communities [1], [2], [3], [4], because it must address the dual challenges of subscription matching in an event space and event dissemination in the network space...
    • ...For example, Aguilera et al. [1] assign subscribers to their closest brokers in the network, ignoring subscriber interests...
    • ...Without loss of generality, assume that B1;���;Bl are the l leaf brokers in B. Each leaf broker Bi is associated with a user-defined capacity fraction �i 2 [0;1], such that P l=1 �i = 1. Perfect load bal-...
    • ...By relaxing the values of Boolean variables to be real numbers (i.e., xij;yik 2 [0;1]), the above mixed integer program can be reduced to an LP. Using an LP algorithm, we compute the optimal fractional solution, and then apply randomized rounding [10] to construct a solution to the filterassignment problem...
    • ...� Closest Broker without Load Balance (Closest: b). This algorithm resembles the one in [1]...
    • ...As discussed in Section I, some previous work considers either subscription similarity in the event space (e.g., [2]) or subscriber location in the network space [1] while ignoring the other aspect...

    Albert Yuet al. Subscriber assignment for wide-area content-based publish/subscribe

    • ...ACM 978-1-4503-0632-4/11/03. on designing ecient content-based indices to enable ecient matching (e.g., [1, 8, 9, 11, 18, 22], and on developing ecient dissemination strategies for delivering the matched events to the subscribers (e.g., [2, 3, 5, 6, 10, 20, 21])...
    • ...This is semantically dierent from the problem of disseminating the results of matches to subscribers in a physically distributed network (e.g., [2, 3, 5, 6, 10, 20, 21])...
    • ...Our proposed solution builds upon the work on indexing content-based subscriptions, including the use of hash-indices (e.g., [11]), trees (e.g., [2]) and inverted lists (e.g., [22]), and extends them to work on graph constraints...

    Andrei Z. Broderet al. Efficiently evaluating graph constraints in content-based publish/subs...

    • ...For example, RT of node 14 is represented by the two intervals [8, 8], and [2, 7]...
    • ...Example 1 We illustrate the retrieval of event B within time interval [2, 8] on the CBS-Tree in Fig. 7b. Initially, at the root, since the largest timestamp represented by node 12 (i.e...
    • ...At first, when (D, 8) is inserted into the tree, the min-latest occurrence retrieval procedure is triggered, since (D, 8) matches a sink for both α1 and α2. We examine α1 first, and find the latest event instance of event B within interval [2, 7]. To achieve this, we load blocks 4 and 3, and then find the answer (B, 7). Next, we try to find the latest event instance of event C within interval [2, 6] for α1. However, we realize that if we ...
    • ...At first, when (D, 8) is inserted into the tree, the min-latest occurrence retrieval procedure is triggered, since (D, 8) matches a sink for both α1 and α2. We examine α1 first, and find the latest event instance of event B within interval [2, 7]. To achieve this, we load blocks 4 and 3, and then find the answer (B, 7). Next, we try to find the latest event instance of event C within interval [2, 6] for α1. However, we realize that if we ...
    • ...If it is, we postpone processing α1 and temporarily save the current state, i.e., finding the latest event instance for event C within interval [2, 6], stopping at node 13, and requiring block 2. We then switch to process α2 and find the latest event instance of event C within interval [2, 7]. After that, block 3 can be replaced by block 2, and thus we can retrieve (C, 6) for α2. Next, similarly, if we want to further retrieve (B, 2) for ...
    • ...If it is, we postpone processing α1 and temporarily save the current state, i.e., finding the latest event instance for event C within interval [2, 6], stopping at node 13, and requiring block 2. We then switch to process α2 and find the latest event instance of event C within interval [2, 7]. After that, block 3 can be replaced by block 2, and thus we can retrieve (C, 6) for α2. Next, similarly, if we want to further retrieve (B, 2) for ...
    • ...At this moment, similarly, we postpone processing α2, temporarily save the current state (finding the latest event instance for event B within interval [2, 4], stopping at node 12, and requiring block 1), and turn back to processing α1. In this way, we can find the min-latest occurrences without frequent flush of the main memory...
    • ...In traditional content-based subscription systems [2,16,18], senders simply publish events (messages or news) associated with a set of predefined attribute-value pairs to the receivers, while receivers subscribe the interested contents of events by specifying individual predicates composed of a set of attribute-value-operator pairs (an example of attributevalue-operator pair can be seen below)...
    • ...Aguilera et al. [2] indexes the subscriptions in a tree structure where a path from the root to a leaf corresponds to a set of subscriptions whose sets of attribute-value-operator triples are identical...
    • ...The query format considered in content-based subscription systems [2,18 ]i s setbased, while ours is graph-based...

    Chung-Wen Choet al. On-line rule matching for event prediction

Sort by: