BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Controlled Data and Resource Sharing  
DTSTART:20100625T101500
DTSTAMP:20260408T233308Z
UID:2d9f8114e46c8eee3f757a91c30dc363034f1981b3545e203456440f
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Sandhya Dwarkadas\, Rochester University\, NY\nTechnolog
 y projections indicate the possibility of 50 billion transistors on a chip
  in a few years\, with the processor landscape being dominated by multi-co
 re designs. Developing correct and reliable software that takes advantage 
 of the multiple cores to deliver high performance\, while at the same time
  ensuring performance isolation\, remains a growing challenge. \n\nIn this
  talk\, I will begin by describing our changes to existing \ncoherence mec
 hanisms in order to control data sharing --- \nin particular\, to monitor 
 memory\, isolate data modifications\, \nsupport fine-grain protection\, an
 d improve the efficiency of fine-grain sharing. \n\nI will discuss the uti
 lity of the mechanisms for a range of applications and show how monitoring
  and isolation can be combined to support one parallel programming constru
 ct --- transactional memory. \n\nThe prevalence of multi-core processors a
 lso implies more ubiquitous \nresource sharing at several levels. Controll
 ed resource sharing \nis essential in order to provide performance isolati
 on\, dictating the need for resource-aware policies within the operating s
 ystem. As time permits\, I will also describe our efforts in combining res
 ource utilization prediction\, resource control mechanisms\, and resource-
 aware policies in order to effect performance isolation at the operating s
 ystem level and improve multi-core resource utilization.   Prof. Dwarkadas
 ' homepage
LOCATION:BC 01 https://plan.epfl.ch/?room==BC%2001
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR