BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260707T115455Z
UID:6ff2c67c-fc26-44ed-9880-550000db509a
DTSTART:20150928T063000Z
DTEND:20151002T143000Z
DESCRIPTION:Distributed memory parallelization with the Message Passing Int
 erface MPI (Mon+Tue\, for beginners – non-PATC part):\nOn clusters and d
 istributed memory architectures\, parallel programming with the Message Pa
 ssing Interface (MPI) is the dominating programming model. The course give
 s an full introduction into MPI-1. Further aspects are domain decompositio
 n\, load balancing\, and debugging. An MPI-3.1 overview and the MPI one-si
 ded communication is also taught. Hands-on sessions (in C and Fortran) wil
 l allow users to immediately test and understand the basic constructs of t
 he Message Passing Interface (MPI).Shared memory parallelization with Open
 MP-4.0 (Wed\, for beginners – non-PATC part):\nThe focus is on shared me
 mory parallelization with OpenMP\, the key concept on hyper-threading\, du
 al-core\, multi-core\, shared memory\, and ccNUMA platforms. This course t
 eaches shared memory OpenMP parallelization. Hands-on sessions (in C and F
 ortran) will allow users to immediately test and understand the directives
  and other interfaces of OpenMP. Race-condition debugging tools are also p
 resented.Advanced topics in parallel programming with MPI-3.1 (Thu+Fri 
 – PATC course):\nTopics are MPI-2 parallel file I/O\, hybrid mixed model
  MPI+OpenMP parallelization\, MPI-3.0 and 3.1\, parallelization of explic
 it and implicit solvers and of particle based applications\, parallel nume
 rics and libraries\, and parallelization with PETSc. MPI-3.0 introduced a 
 new shared memory programming interface\, which can be combined with MPI m
 essage passing and remote memory access on the cluster interconnect. It ca
 n be used for direct neighbor accesses similar to OpenMP or for direct hal
 o copies\, and enables new hybrid programming models. These models are com
 pared in the hybrid mixed model MPI+OpenMP parallelization session with va
 rious hybrid MPI+OpenMP approaches and pure MPI.  Hands-on sessions are i
 ncluded. Only this 3rd part is sponsored by the PATC program.\n\nIf you wa
 nt to register to the whole course or to more than the last two days\, or
  if you are a non-academic participant\, or participant from outside th
 e PRACE-member-countries\, then please register through the HLRS course pa
 ge.\n\n \n\nhttps://events.prace-ri.eu/event/345/
SUMMARY:Parallel Programming with MPI and OpenMP and Advanced Parallel Prog
 ramming @ HLRS
URL;VALUE=URI:https://events.prace-ri.eu/event/345/
END:VEVENT
END:VCALENDAR