BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260705T092150Z
UID:3c9e2266-2b85-49dd-a26b-9d96e540204a
DTSTART:20171130T080000Z
DTEND:20171201T160000Z
DESCRIPTION:This course teaches performance engineering approaches on the c
 ompute node level. "Performance engineering" as we define it is more than 
 employing tools to identify hotspots and bottlenecks. It is about developi
 ng a thorough understanding of the interactions between software and hardw
 are. This process must start at the core\, socket\, and node level\, where
  the code gets executed that does the actual computational work. Once the 
 architectural requirements of a code are understood and correlated with pe
 rformance measurements\, the potential benefit of optimizations can often 
 be predicted. We introduce a "holistic" node-level performance engineering
  strategy\, apply it to different algorithms from computational science\, 
 and also show how an awareness of the performance features of an applicati
 on may lead to notable reductions in power consumption.\n\nThe course is a
  PRACE Advanced Training Center event.\n\nIntroduction\n	Our approach to p
 erformance engineering\n		Basic architecture of multicore systems: threads
 \, cores\, caches\, sockets\, memory\n		The important role of system topol
 ogy\n	\n	Tools: topology &amp\; affinity in multicore environments\n	Overv
 iew\n		likwid-topology and likwid-pin\n	\n	Microbenchmarking for architect
 ural exploration\n	Properties of data paths in the memory hierarchy\n		Bot
 tlenecks\n		OpenMP barrier overhead\n	\n	Roofline model: basics\n	Model as
 sumptions and construction\n		Simple examples\n		Limitations of the Roofli
 ne model\n	\n	Pattern-based performance engineering\n	Optimal use of paral
 lel resources\n	Single Instruction Multiple Data (SIMD)\n		Cache-coherent 
 Non-Uniform Memory Architecture (ccNUMA)\n		Simultaneous Multi-Threading (
 SMT)\n	\n	Tools: hardware performance counters\n	Why hardware performance 
 counters?\n		likwid-perfctr\n		Validating performance models\n	\n	Roofline
  case studies\n	Dense matrix-vector multiplication\n		Sparse matrix-vector
  multiplication\n		Jacobi (stencil) smoother\n	\n	Optional: The ECM perfor
 mance model\n\n\nhttps://events.prace-ri.eu/event/632/
SUMMARY:Node-Level Performance Engineering @ LRZ
URL;VALUE=URI:https://events.prace-ri.eu/event/632/
END:VEVENT
END:VCALENDAR