BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:IRONModel: robust performance models in the wild - Eno Thereska (M
 icrosoft Research Cambridge)
DTSTART:20090219T160000Z
DTEND:20090219T170000Z
UID:TALK16757@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:We look into completely modelling a storage-based data centre 
 and find that traditional performance models are too brittle. Simply put\,
  a brittle model is often inaccurate and incorrect. That turns off systems
  people and administrators who then opt for simple\, but often inaccurate\
 , "rules-of-thumb". After learning lessons the hard way\, we went and impl
 emented a robust modelling framework we call IRONModel. IRONModel leverage
 s the redundancy of high-level system specifications described through (qu
 eueing) models and low-level system implementation to localize many types 
 of system-model inconsistencies. IRONModel can guide designers to the pote
 ntial source of the discrepancy\, and\, if appropriate\, can semi-automati
 cally evolve the models (using statistical techniques) to handle unanticip
 ated inputs.\nThis is largely a systems talk and touches on the minimal ma
 thematical tools needed to build robust models within it\, which then help
  manage the system better. This is joint work with CMU.\n\nBio: I am a sys
 tems person\, currently with a focus on file systems and storage technolog
 ies and high-performance data centers. I also have great interest in apply
 ing machine learning and queuing analysis to help simplify and automate sy
 stem management. Since September 2007 I have been a Researcher at Microsof
 t Research in Cambridge\, UK. I received my PhD/MS/BS from Carnegie Mellon
 .\n
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR