Models and issues in data stream systems
Proceedings of the twenty-first ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
Adaptive filters for continuous queries over distributed data streams
Proceedings of the 2003 ACM SIGMOD international conference on Management of data
Gigascope: a stream database for network applications
Proceedings of the 2003 ACM SIGMOD international conference on Management of data
Adaptive stream resource management using Kalman Filters
SIGMOD '04 Proceedings of the 2004 ACM SIGMOD international conference on Management of data
Predictive filtering: a learning-based approach to data stream filtering
DMSN '04 Proceeedings of the 1st international workshop on Data management for sensor networks: in conjunction with VLDB 2004
Sketching streams through the net: distributed approximate query tracking
VLDB '05 Proceedings of the 31st international conference on Very large data bases
Processing of Data Streams with Prediction Functions
HICSS '06 Proceedings of the 39th Annual Hawaii International Conference on System Sciences - Volume 09
Distributed set-expression cardinality estimation
VLDB '04 Proceedings of the Thirtieth international conference on Very large data bases - Volume 30
Research on cost-efficient processing of continuous extreme queires
IITA'09 Proceedings of the 3rd international conference on Intelligent information technology application
Data stream forecasting for system fault prediction
Computers and Industrial Engineering
| Hi-index | 0.00 |
A framework is presented to provide a mechanism to maintain adaptive prediction models established both on the coordinator and remote nodes in distributed data stream processing for reducing communication consumption. The coordinator employs these models to answer registered queries, while the remote nodes check whether the prediction value is close to the actual value or not. Update messages are needed only when there's a large deviation between prediction value and actual value. Three particular prediction models are given and compared with existent ones. Analytical and experimental evidence show that the proposed approach performs better both on overall communication cost reduction and prediction query processing.