DMCA
End-to-End Support for Joins in Large-Scale Publish/Subscribe Systems
| Citations: | 14 - 1 self |
BibTeX
@MISC{Chandramouli_end-to-endsupport,
author = {Badrish Chandramouli and Jun Yang},
title = {End-to-End Support for Joins in Large-Scale Publish/Subscribe Systems},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We address the problem of supporting a large number of select-join subscriptions in a wide-area publish/subscribe system. Subscriptions are interested in joins over different data sources (tables), with varying interests expressed as range selection predicates over table attributes. Naive schemes, such as computing and sending join results from a server, are inefficient because they produce redundant data, and are unable to share dissemination costs across subscribers and events. We propose a novel, scalable scheme that group-processes and disseminates a general mix of multi-way select-join subscriptions. We also propose a simple and application-agnostic extension tocontent-drivennetworks (CN), which further improves sharing of dissemination costs across events and subscribers. We develop and experimentally evaluate our scheme, and show that it can generate an order of magnitude lower network traffic at very low processing cost. Our extension to CN can further reduce traffic by another order of magnitude, with almost no increase in notification latency. 1
Keyphrases
end-to-end support large-scale publish subscribe system range selection multi-way select-join subscription application-agnostic extension tocontent-drivennetworks select-join subscription different data source scalable scheme large number share dissemination cost low processing cost join result dissemination cost network traffic table attribute wide-area publish subscribe system notification latency naive scheme general mix redundant data
