BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Acceleration of scientific computing using graphics hardware\n - G
 raham Pullan\,  Department of Engineering\, University of Cambridge
DTSTART:20080528T131500Z
DTEND:20080528T141500Z
UID:TALK11960@talks.cam.ac.uk
CONTACT:Timothy G. Griffin
DESCRIPTION:Developers of scientific computing codes crave performance (FL
 OPS) but only a tiny minority can afford to commission bespoke hardware. A
 s a result\, commodity hardware is co-opted into the HPC role. Most scient
 ific applications are inherently data parallel and\, fortunately\, the sta
 gnation in CPU clock speed has coincided with cheap high performance netwo
 rking infrastructure so that PC clusters are now ubiquitous. Another way t
 o exploit parallelism is through multi-core shared memory devices and comm
 odity PCs with 2\, 4 (8...) cores are now commonplace. However\, equally p
 revalent\, and far more powerful\, are graphics processing units (GPUs). G
 PUs typically have over 100 cores and generate an order of magnitude great
 er floating point performance than CPUs. This seminar introduces the GPU a
 s a science co-processor with fluid dynamics simulations as the example ap
 plication.
LOCATION:Lecture Theatre 1\, Computer Laboratory
END:VEVENT
END:VCALENDAR