BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Condensing the cloud: running CIEL on many-core - Malte Schwarzkop
 f (University of Cambridge)
DTSTART:20110407T150000Z
DTEND:20110407T160000Z
UID:TALK30733@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:Distributed execution engines have revolutionized data process
 ing by making parallel programming simple. Systems such as MapReduce\, Dry
 ad and Hadoop can achieve massive throughput when running on thousands of 
 commodity servers\, yet generally only require the programmer to provide s
 equential code. These systems were designed to scale out across many worke
 r machines\, each of which had at most a handful of processors. As such\, 
 intra-server parallelism is either left entirely to the developer or manag
 ed centrally. In this talk\, I will compare the performance of our recentl
 y-developed CIEL distributed execution engine on three quite different 48-
 core platforms: an AMD "Magny-Cours" ccNUMA server\, an experimental Intel
  Single-Chip Cloud computer\, and an Amazon EC2 cluster of 48 uniprocessor
  VMs. Our results demonstrate task creation/coordination overhead to be a 
 problem when using fine-grained tasks\, and we demonstrate a few simple im
 provements that mitigate these issues. We also outline a set of further ch
 allenges and opportunities.\n\nThis is a practice talk for SFMA 2011 at Eu
 rosys.\n\nPaper: http://www.cl.cam.ac.uk/~ms705/pub/papers/2011-ciel-sfma.
 pdf\n\nSFMA: http://research.microsoft.com/en-us/um/people/rmcilroy/SFMA/i
 ndex.htm\n\n
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR