Abstract | ||
---|---|---|
The workload in a database system encompasses cluster of multiple queries running concurrently. The requirement of business is that the workload which consists of different mixes of queries should complete within a short period. We propose a scheduler called Gscheduler, which schedules queries to form good queries mixes in order to finish the workload quickly. The rationale is that a query mix consisting of multiple queries that interact each other and the interactions can significantly delay or accelerate the execution of the mix. We propose a notion called mix rating to measure query interactions in a mix, which is used to differentiate good mixes from bad mixes. Experimental results show the effectiveness of the scheduler. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1007/978-3-030-02934-0_36 | WISA |
Field | DocType | Citations |
Computer science,Workload,Theoretical computer science,Schedule,Performance management,Database,Workload management | Conference | 0 |
PageRank | References | Authors |
0.34 | 12 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Muhammad Amjad | 1 | 15 | 1.29 |
Jinwen Zhang | 2 | 0 | 0.68 |