BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Rethinking GPU Execution Model
DTSTART:20181115T111500
DTEND:20181115T121500
DTSTAMP:20260410T034213Z
UID:f0de41b9e52ecfe0e0a1c969659ac6186fb4f268f51be053eb63735a
CATEGORIES:Conferences - Seminars
DESCRIPTION:Yunho Oh\nGraphics processing units (GPUs) have become the arc
 hitectural choice to achieve high throughput in general-purpose computing.
  Thread-level parallelism (TLP) in GPUs is implemented by concurrently exe
 cuting a large number of threads. However\, GPUs cannot often achieve the 
 theoretical peak performance. I found that the critical performance bottle
 necks on GPUs are 1) limited memory system performance and 2) limited thre
 ad scheduling resources and register file.    In this talk\, I will show
  the GPU execution model and two above performance bottlenecks on GPUs in 
 detail. Then\, I will introduce two solutions addressing these challenges.
  First\, I will introduce a new GPU architecture\, called Adaptive PREfetc
 hing and Scheduling (APRES)\, that overcomes the limited memory system per
 formance by improving cache efficiency on GPUs. Second\, I will introduce 
 another work\, called FineReg\, that provides a solution to schedule threa
 ds over the limits of scheduling resources and register file on GPUs.
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR