The Query Planning for Continuous Aggregation and Queries Using Data Aggregators

Abstract:

Continuous aggregation queries are used to monitor the changes in data with time varying for online decision making. For continuous queries low cost and scalable techniques used a network of aggregators. Individual node cannot by itself determine its inclusion in the query result for this different algorithmic challenges from aggregate and selection queries are presented. At specific coherencies each data item can serve for a set of data aggregators. Technique involves disseminating query into sub query and sub queries are executed on the chosen data aggregators. We build a query cost model which can be used to estimate the number of refresh messages which is required to satisfy the client specified incoherency bound. Performance results shows that by our method the query can be executed using less than one third the messages required for existing schemes. Our adaptive strategy employs distributed decisions made by the distributed servers independently based on localized statistics collected by each server at runtime. When comparatively static environment, propose two motionless tree construction algorithms relying on apriori system statistics. These static trees can also be used as early trees in a dynamic environment and apply our schemes to both single  and multi object distribution. Our extensive performance study illustrate that the adaptive mechanisms.

Index Terms — Algorithms, Continuous Queries, Data Dissemination, Distributed Query Processing, Coherency, Performance.