Load Shedding for Shared Window Join over Real-Time Data Streams

Authors:
Li Ma;Dangwei Liang;Qiongsheng Zhang;Xin Li;Hongan Wang
Affiliations:
School of Computer Science and Communication Engineering, China University of Petroleum, Dongying, P.R. China 257061 and Institute of Software, Chinese Academy of Sciences, Beijing, P.R. China 100 ...;Geophysical Research Institute of Shengli Oil Field, China Petroleum & Chemical Corporation, Dongying, P.R. China 257000;School of Computer Science and Communication Engineering, China University of Petroleum, Dongying, P.R. China 257061;Shandong University, Ji'nan, P.R. China 250101;Institute of Software, Chinese Academy of Sciences, Beijing, P.R. China 100190
Venue:
APWeb/WAIM '09 Proceedings of the Joint International Conferences on Advances in Data and Web Management
Year:
2009

Citing 11
Cited 0

Continuously adaptive continuous queries over streams

Proceedings of the 2002 ACM SIGMOD international conference on Management of data
Issues in data stream management

ACM SIGMOD Record
Load Shedding for Aggregation Queries over Data Streams

ICDE '04 Proceedings of the 20th International Conference on Data Engineering
Adaptive load shedding for windowed stream joins

Proceedings of the 14th ACM international conference on Information and knowledge management
Approximating StreamingWindow Joins Under CPU Limitations

ICDE '06 Proceedings of the 22nd International Conference on Data Engineering
Load shedding in stream databases: a control-based approach

VLDB '06 Proceedings of the 32nd international conference on Very large data bases
Prediction-Based QoS Management for Real-Time Data Streams

RTSS '06 Proceedings of the 27th IEEE International Real-Time Systems Symposium
Adaptive Load Management over Real-Time Data Streams

FSKD '07 Proceedings of the Fourth International Conference on Fuzzy Systems and Knowledge Discovery - Volume 02
Scheduling for shared window joins over data streams

VLDB '03 Proceedings of the 29th international conference on Very large data bases - Volume 29
Load shedding in a data stream manager

VLDB '03 Proceedings of the 29th international conference on Very large data bases - Volume 29
Memory-limited execution of windowed stream joins

VLDB '04 Proceedings of the Thirtieth international conference on Very large data bases - Volume 30

Quantified Score

Hi-index	0.00

Visualization

Abstract

Join is a fundamental operator in a Data Stream Management System (DSMS). It is more efficient to share execution of multiple windowed joins than separate execution of everyone because the former saves a part of cost in common windows. Therefore, shared window join is adopted widely in multi-queries DSMS. When all tasks of queries exceed maximum system capacity, the overloaded DSMS fails to process all of its input data and keep up with the rates of data arrival. Especially in a time-critical environment, queries should be completed not just timely but within certain deadlines. In this paper, we address load shedding approach for shared window join over real-time data streams. A load shedding algorithm LS-SJRT-CW is proposed to handle queries shared window join in overloaded real-time system effectively. It would reduce load shedding overhead by adjusting sliding window size. Experiment results show that our algorithm would decrease average deadline miss ratio over some ranges of workloads.