BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:An Experiment-driven Approach to Tune Database Configuration Param
 eters and SQL
DTSTART:20090514T133000
DTSTAMP:20260407T025915Z
UID:3b3901ce76f831afbf849787fd4be7dfbc88a99d6d9b847093eeda3a
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Shivnath BABU\, Duke University\nAbstract\nDatabase syst
 ems have a large number of configuration parameters that control memory di
 stribution\, I/O optimization\, costing of query plans\, and other behavio
 r. Regular users and even expert database administrators struggle to tune 
 these parameters for good performance.  In the first part of the talk\, I 
 will present iTuned\, a tool that automates the task of identifying good s
 ettings for database configuration parameters.  iTuned has three novel fea
 tures: (i) a technique called Adaptive Sampling that collects monitoring d
 ata through planned experiments to find high-impact parameters and high-pe
 rformance parameter settings\, (ii) support for live experiments in produc
 tion database environments with little overhead on the user-facing workloa
 d\;  and (iii)  portability across different database systems.\n\nThe seco
 nd part of the talk will consider SQL tuning\, the attempt to improve a po
 orly-performing execution plan selected by the query optimizer for a SQL q
 uery. Like parameter tuning\, SQL tuning requires considerable expertise i
 n database internals\; forcing database administrators to rely on intuitio
 n or trial-and-error. I will show how SQL tuning can be automated using th
 e same experiment-driven paradigm as iTuned\, but with a different instant
 iation. Finally\, I will present evidence from empirical evaluations and r
 ecent computing trends to argue why the experiment-driven paradigm is an i
 dea whose time has come.\n\nSpeaker's bio\nShivnath Babu is an Assistant P
 rofessor of Computer Science at Duke University. He got his Ph.D. from Sta
 nford University in 2005. He has received a U.S. National Science Foundati
 on CAREER Award and three IBM Faculty Awards. His current research focuses
  on making large-scale databases and systems easier to manage.
LOCATION:BC 03 https://plan.epfl.ch/?room==BC%2003
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR