BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:The CAVA Computer:  Exceptional Parallelism and Energy Efficiency
DTSTART:20150713T110000
DTEND:20150713T120000
DTSTAMP:20260407T020953Z
UID:39df0c5536d4cd20f3bfb39aa9a1c0a994d47eb6ff7630239dadd3a3
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Peter Hsu\, Oracle\nPeter Hsu was born in Hong Kong and ca
 me to the United States at age 15.  He received a B.S. degree from the Un
 iversity of Minnesota at Minneapolis in 1979\, and the M.S. and Ph.D. degr
 ees from the University of Illinois at Urbana-Champaign in 1983 and 1985\,
  respectively\, all in Computer Science.  His first job was at IBM Resear
 ch in Yorktown Heights from 1985-1987\, working on superscalar code genera
 tion with the 801 compiler team.  He then joined his ex-professor at Cydr
 ome\, which developed an innovative VLIW computer.  In 1988 he moved to S
 un Microsystems and tried to build a water-cooled gallium arsenide SPARC p
 rocessor\, but the technology was not sufficiently mature and the effort f
 ailed.  He joined Silicon Graphics in 1990 and designed the MIPS R8000 TF
 P microprocessor\, which shipped in the SGI Power Challenge systems in 199
 5.  He became a Director of Engineering until 1997\, then left to co-foun
 d his own startup\, ArtX\, best known for designing the Nintendo GameCube.
   ArtX was acquired by ATI Technologies in 2000\, which has since been ac
 quired by AMD.  Peter left ArtX in 1999 and worked briefly at Toshiba Ame
 rica\, then became a visiting Industrial Researcher at the University of W
 isconsin at Madison in 2001.  He then consulted part time at various star
 tups\, and attended the Art Academy University and the California College 
 of the Arts in San Francisco where he learned to paint oil portraits\, and
  a Paul Mitchell school where he learned to cut and color hair.  In the l
 ate 2000’s he consulted for Sun Labs\, which lead to discussions about t
 he RAPID project post acquisition.  Peter joined Oracle Labs as an Archit
 ect in 2011.\nAbstract :\nDesigning a new computer system is a very expens
 ive proposition. But 80% of it is exactly the same as every other computer
 —you need caches\, multiprocessing\, coherency protocols\, memory system
 s\, etc.  Getting all of that right requires skill and experience\, but i
 s taken for granted and does not command much of a premium.  Getting it w
 rong\, on the other hand\, is a commercial disaster.  In this talk I prop
 ose a research initiative to standardize and open-source a design for the 
 80% that is the same in every design\, so that everyone can concentrate on
  adding value to their own remaining 20%.  The CAVA computer is a “clus
 ter in a rack” energy-efficient parallel computing architecture targetin
 g 10nm CMOS technology.  The first part of the talk describes a 1024-node
  system where each node consists of 96-core\, 3-issue out-of-order process
 or chips running at 1GHz with four DDR4 memory channels.  Power estimates
  of different components are discussed\, as well as cost projections.  Th
 e second part of the talk discusses architectural tradeoffs that were made
 \, how this architecture might play in the HPC exa-scale arena\, and broad
 er market implications.  The talk concludes with how I envision the simul
 ation infrastructure is organized\, what Oracle Labs brings to the table\,
  and a list of research topics that I and others at Oracle Labs are active
 ly researching and would be interested in working with students at Univers
 ities.  I hope to organize an in-depth research effort into designing thi
 s computer and\, if sufficient progress can be made\, perhaps building a p
 rototype.
LOCATION:INF 211 http://plan.epfl.ch/?zoom=19&recenter_y=5863880.47179&rec
 enter_x=730652.58335&layerNodes=fonds\,batiments\,labels\,information\,par
 kings_publics\,arrets_metro\,transports_publics&floor=2&q=INF211
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR