Adaptive Class-Based Scheduling of Continuous Queries

The emergence of Data Stream Management Systems (DSMS) facilitates implementing many types of monitoring applications via continuous queries (CQs). However, these applications usually have different quality-of-service requirements for different CQs. In this work, we are proposing the Adaptive Broadcast Disks (ABD) scheduler, a new scheduling policy which employs two-level scheduling that can handle different ranks of CQ classes. The ABD scheduler optimizes the weighted average response time of the CQ classes while still preserving the relative importance of each class. We demonstrate that ABD outperforms state-of-the-art schedulers and adapts to changes in the workload without manual intervention.