Hybrid content-based publish/subscribe system
Qing Zhou, Xiong Zhiyong, Rong Hong, Tao You, Nan Jiang
Abstract
Qing Zhou, Xiong Zhiyong, Rong Hong, Tao You, Nan Jiang
Abstract
Current typical content-based publish/subscribe systems (CPS) are not efficient in subscription processing or event matching. This paper presents hybrid content-based publish/subscribe system (HCPS), a novel CPS which joins predicate indexing and testing network approaches. We put subscription coverage mechanism, which are separated from testing network structures, into predicate indexing mechanism to sustain efficient matching, whilst adding, deleting and changing large number of subscriptions. Finally, experiments and performance analysis show that HCPS significantly improve throughput of event propagation and reduce response time to subscription updates.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Current typical content-based publish/subscribe systems (CPS) are not efficient in subscription processing or event matching. This paper presents hybrid content-based publish/subscribe system (HCPS), a novel CPS which joins predicate indexing and testing network approaches. We put subscription coverage mechanism, which are separated from testing network structures, into predicate indexing mechanism to sustain efficient matching, whilst adding, deleting and changing large number of subscriptions. Finally, experiments and performance analysis show that HCPS significantly improve throughput of event propagation and reduce response time to subscription updates.
Key concepts: Computer science, Joins, Search engine indexing, Publication, Matching (statistics), Predicate (mathematical logic), Event (particle physics), Computer network