BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Improving Content Delivery Networks by Tracking Geographic Social 
 Cascades - Salvatore Scellato (University of Cambridge)
DTSTART:20110310T160000Z
DTEND:20110310T170000Z
UID:TALK29351@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:Providers such as YouTube o er easy access to multimedia conte
 nt to millions\, generating high bandwidth and storage demand on the Conte
 nt Delivery Networks they rely upon.\nMore and more\, the di usion of this
  content happens on online social networks such as Facebook and Twitter\, 
 where social cascades can be observed when users increasingly re- post lin
 ks they have received from others.\nIn this paper we describe how geograph
 ic information ex- tracted from social cascades can be exploited to improv
 e caching of multimedia les in a Content Delivery Network.\nWe take advant
 age of the fact that social cascades can prop- agate in a geographically l
 imited area to discern whether an item is spreading locally or globally. T
 his informs cache re- placement policies\, which utilize this information 
 to ensure that content relevant to a cascade is kept close to the users wh
 o may be interested in it.\nWe validate our approach by using a novel data
 set which combines social interaction data with geographic information: we
  track social cascades of YouTube links over Twitter and build a proof-of-
 concept geographic model of a realis- tic distributed Content Delivery Net
 work. Our performance evaluation shows that we are able to improve cache h
 its with respect to cache policies without geographic and social in- forma
 tion.\n\n
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR