Premium
Improving scalability of event‐driven distributed objects architectures
Author(s) -
Mencnarowski Daniel,
Zieliński Krzysztof
Publication year - 2000
Publication title -
software: practice and experience
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.437
H-Index - 70
eISSN - 1097-024X
pISSN - 0038-0644
DOI - 10.1002/1097-024x(20001110)30:13<1509::aid-spe349>3.0.co;2-t
Subject(s) - scalability , computer science , middleware (distributed applications) , distributed computing , multicast , common object request broker architecture , event (particle physics) , quality of service , service (business) , routing (electronic design automation) , computer network , database , physics , economy , quantum mechanics , economics
The goal of this paper is to present practical techniques for improving the scalability of a distributed objects architecture built around an event‐driven paradigm and the evaluation of their effectiveness in the case of location systems. As a case study, the ABng location system constructed using a publisher/subscriber pattern over a CORBA compliant middleware is presented. The main contribution of this paper is the elaboration and evaluation of three main concepts that play a crucial role in improving the scalability of event‐driven architectures: events dissemination quality of service, notification tree based events‐routing, and transparent multicast event‐channel notification service. The applicability and efficiency of these techniques have been verified in practice by the ABng system implementation and performance testing. Copyright © 2000 John Wiley & Sons, Ltd.