Paper Info
Title
Multi-query SQL progress indicators
Abstract
Recently, progress indicators have been proposed for SQL queries in RDBMSs. All previously proposed progress indicators consider each query in isolation, ignoring the impact simultaneously running queries have on each other’s performance. In this paper, we explore a multi-query progress indicator, which explicitly considers concurrently running queries and even queries predicted to arrive in the future when producing its estimates. We demonstrate that multi-query progress indicators can provide more accurate estimates than single-query progress indicators. Moreover, we extend the use of progress indicators beyond being a GUI tool and show how to apply multi-query progress indicators to workload management. We report on an initial implementation of a multi-query progress indicator in PostgreSQL and experiments with its use both for estimating remaining query execution time and for workload management.
Year
DOI
Venue
2006
10.1007/11687238_54
EDBT
Keywords
Field
DocType
accurate estimate,remaining query execution time,single-query progress indicator,multi-query progress indicator,multi-query sql progress indicator,workload management,sql query,progress indicator,initial implementation,gui tool
SQL,Data mining,Zipf's law,Database query,Workload,Computer science,Time management,Execution time,User interface,Database,Workload management
Conference
Volume
ISSN
ISBN
3896
0302-9743
3-540-32960-9
Citations 
PageRank 
References 
36
2.41
13
Authors
3
Name
Order
Citations
PageRank
Gang Luo174144.73
Jeffrey F. Naughton283631913.71
Philip S. Yu3306703474.16