BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:High-Performance Processing with Field-Programmable Logic - George
  Constantinides\, Imperial College London
DTSTART:20080220T141500Z
DTEND:20080220T151500Z
UID:TALK9621@talks.cam.ac.uk
CONTACT:Timothy G. Griffin
DESCRIPTION:Field-programmable logic devices\, such as Field-Programmable 
 Gate Arrays (FPGAs) are ‘programmable hardware’\, and may be configure
 d at startup or on-the-fly to implement a wide variety of digital electron
 ic designs. These devices are commonly used for rapid prototyping of compl
 ex computational systems. However\, I will discuss a different paradigm 
 – the use of FPGAs as general purpose computers\, or as part of a genera
 l purpose computer system. In this approach\, conceived by multiple resear
 chers over the last 10-15 years\, but only now coming of age\, the general
  purpose FPGA is configured to act as a special purpose computational mach
 ine for whatever computation is required at the present time\; the archite
 cture is designed around the algorithm.\n\n \n\nI will briefly discuss the
  evolution of FPGA architectures and draw comparisons with the evolution o
 f general purpose processors\, before discussing two main advantages of mo
 dern FPGA architectures\, the ability to design numerical representation a
 round the algorithm\, and the potential memory bandwidth available.\n\n \n
 \nAlthough companies such as Cray and SRC now incorporate FPGAs in their n
 ew supercomputers\, one of the main challenges remains how to program such
  devices. We will touch on this question via an example providing food for
  thought. Finally\, I will present some numerical comparisons of performan
 ce achieved using FPGAs compared to traditional general purpose processors
 \, showing 1-2 orders of magnitude improvements in speed for particular ap
 plications.
LOCATION:Lecture Theatre 1\, Computer Laboratory
END:VEVENT
END:VCALENDAR