BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260627T145217Z
UID:758b6ca6-8e42-4712-8acd-da14329bd9d2
DTSTART:20140522T073000Z
DTEND:20140523T153000Z
DESCRIPTION:Keywords\nRuntime systems\, thread programming\, accelerators\,
  hardware abstraction\, performance portability.Course description:\nThis 
 course will present the state of the art of runtime system support for pro
 gramming heterogeneous platforms. Heterogeneous computing platforms—such
  as multicores equipped with accelerators—are notoriously difficult to p
 rogram due to the strong differences in performance characteristics among 
 the various available computing units and also to the discrete memory spac
 es of accelerating boards.The course will be based on two specific runtime
  systems developed at Inria: StarPU\, developed by the Runtime Team in Bor
 deaux\, and XKAAPI\, from the Moais team in Grenoble. It will also present
  the hardware locality library hwloc for discovering hardware resources an
 d the Ezperf framework for solving performance issues.Learning outcomes\nP
 articipants will understand the benefit of the task-based programming mode
 l together with performance modeling\, automatic data management and dynam
 ic scheduling in speeding up application development on heterogeneous comp
 uting platforms and providing long term performance portability.Prerequisi
 tes\nBasic knowledge of C programming language and accelerator programming
  languages (nVidia Cuda\, OpenCL). \n\nhttps://events.prace-ri.eu/event/26
 9/
SUMMARY:Runtime systems for heterogeneous platform programming@ Inria
URL;VALUE=URI:https://events.prace-ri.eu/event/269/
END:VEVENT
END:VCALENDAR