Abstract | ||
---|---|---|
It is challenging to process real-time data service requests, such as online trade or traffic monitoring requests, within their deadlines, while providing differentiated real-time data services. To address the problem, we present new approaches to 1) assigning deadlines to real-time data service requests based on their data access needs and service classes to differentiate real-time data services when the system is busy and 2) closely supporting the specified target delay to deadline ratio (DDR)—the ratio of actual data service delays to deadlines—via feedback control even in the presence of dynamic workloads. Further, we have actually implemented our approaches by extending an open source database unlike most existing work on real-time databases. The experimental results show that our approach closely supports the target DDR bound and service differentiation requirements among the service classes unlike the tested baselines representing the current state of the art in real-time database research. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/TKDE.2015.2441725 | IEEE Transactions on Knowledge and Data Engineering |
Keywords | Field | DocType |
Real-time systems,Telecommunication traffic,Databases,Feedback control,Control systems,Monitoring,Online services,Stock markets | Real-time data,Computer science,Baseline (configuration management),Control system,Data as a service,Data access,Distributed computing | Journal |
Volume | Issue | ISSN |
27 | 12 | 1041-4347 |
Citations | PageRank | References |
0 | 0.34 | 29 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Y. Zhou | 1 | 163 | 37.69 |
Kyoung-Don Kang | 2 | 563 | 37.51 |