BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260707T164058Z
UID:5a3513c0-9269-44e4-af1e-6747d3e395b3
DTSTART:20190220T080000Z
DTEND:20190221T160000Z
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 and apply it to different algorithms from computational science.
  Architectural details that are relevant for performance\, such as pipelin
 ing\, SIMD\, superscalarity\, memory hierarchies\, etc.\, are covered in d
 ue detail.\n\nThe course is a PRACE training event.\n\n\n	Introduction\n	\
 n		Our approach to performance engineering\n		Basic architecture of multic
 ore systems: threads\, cores\, caches\, sockets\, memory\n		The important 
 role of system topology\n	\n	\n	Tools: topology &amp\; affinity in multico
 re environments\n	\n		Overview\n		likwid-topology and likwid-pin\n	\n	\n	M
 icrobenchmarking for architectural exploration\n	\n		Properties of data pa
 ths in the memory hierarchy\n		Bottlenecks\n		OpenMP barrier overhead\n	\n
 	\n	Roofline model: basics\n	\n		Model assumptions and construction\n		Sim
 ple examples\n		Limitations of the Roofline model\n	\n	\n	Pattern-based pe
 rformance engineering\n	Optimal use of parallel resources\n	\n		Single Ins
 truction Multiple Data (SIMD)\n		Cache-coherent Non-Uniform Memory Archite
 cture (ccNUMA)\n		Simultaneous Multi-Threading (SMT)\n	\n	\n	Tools: hardwa
 re performance counters\n	\n		Why hardware performance counters?\n		likwid
 -perfctr\n		Validating performance models\n	\n	\n	Roofline case studies\n	
 \n		Dense matrix-vector multiplication\n		Sparse matrix-vector multiplicat
 ion\n		Jacobi (stencil) smoother\n	\n	\n	Optional: The ECM performance mod
 el\n\nhttps://events.prace-ri.eu/event/821/
SUMMARY:Node-Level Performance Engineering @ LRZ
URL;VALUE=URI:https://events.prace-ri.eu/event/821/
END:VEVENT
END:VCALENDAR