Show simple item record

dc.contributor.authorGruska, Natalieen
dc.date2011-06-09 11:02:31.595
dc.date.accessioned2011-06-09T20:46:56Z
dc.date.available2011-06-09T20:46:56Z
dc.date.issued2011-06-09T20:46:56Z
dc.identifier.urihttp://hdl.handle.net/1974/6545
dc.descriptionThesis (Master, Computing) -- Queen's University, 2011-06-09 11:02:31.595en
dc.description.abstractDatabase Management Systems (DBMSs) are an integral part of many applications. Web-based applications, such as e-commerce sites, are faced with highly variable workloads. The number of customers browsing and purchasing items varies throughout the day and business managers can further complicate the workload by requesting complex reports on sales data. This means the load on a database system can fluctuate dramatically with a sudden influx of requests or a request involving a complex query. If there are too many requests operating in the DBMS concurrently, then resources are strained and performance drops. To keep the DBMS’s performance consistent across varying loads, a load control system can be used. This thesis investigates the concept of a load control system based on regulating individual resource usage in a predictive manner. For the purpose of this proof-of- concept study, we focus on a specific resource; namely, the sort heap. A method of estimating sort heap usage based on the query execution plan is presented and several scheduling methods based on these estimations are proposed. A prototype load control system is used to evaluate and compare the scheduling methods. Experiments show that it is possible to both estimate sort heap requirements and to control sort heap usage using our load control system.en
dc.language.isoengen
dc.relation.ispartofseriesCanadian thesesen
dc.rightsThis publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.en
dc.subjectcomputer scienceen
dc.subjectdatabase management systemsen
dc.subjectquery schedulingen
dc.titleResource-Aware Query Scheduling in Database Management Systemsen
dc.typethesisen
dc.description.degreeM.Sc.en
dc.contributor.supervisorMartin, Patricken
dc.contributor.departmentComputingen
dc.degree.grantorQueen's University at Kingstonen


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record