BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20260707T100901Z
UID:4188219b-a632-4c3d-b090-b1f92ee4eea3
DTSTART:20200514T090000Z
DTEND:20200522T160000Z
DESCRIPTION:The world’s largest supercomputers are used almost exclusivel
 y to run applications which are parallelised using Message Passing. The co
 urse covers all the basic knowledge required to write parallel programs us
 ing this programming model\, and is directly applicable to almost every pa
 rallel computer architecture.\n\nParallel programming by definition involv
 es co-operation between processes to solve a common task. The programmer h
 as to define the tasks that will be executed by the processors\, and also 
 how these tasks are to synchronise and exchange data with one another. In 
 the message-passing model the tasks are separate processes that communicat
 e and synchronise by explicitly sending each other messages. All these par
 allel operations are performed via calls to some message-passing interface
  that is entirely responsible for interfacing with the physical communicat
 ion network linking the actual processors together. This course uses the d
 e facto standard for message passing\, the Message Passing Interface (MPI)
 . It covers point-to-point communication\, non-blocking operations\, deriv
 ed datatypes\, virtual topologies\, collective communication and general d
 esign issues.\n\nThe course is normally delivered in an intensive three-da
 y format using EPCC’s dedicated training facilities. It is taught using 
 a variety of methods including formal lectures\, practical exercises\, pro
 gramming examples and informal tutorial discussions. This enables lecture 
 material to be supported by the tutored practical sessions in order to rei
 nforce the key concepts.\n\nThis course is free to all academics. \n\nInt
 ended Learning Outcomes\n\nOn completion of this course students should be
  able to: Understand the message-passing model in detail. Implement standa
 rd message-passing algorithms in MPI. Debug simple MPI codes. Measure and 
 comment on the performance of MPI codes. Design and implement efficient pa
 rallel programs to solve regular-grid problems.\n\nPre-requisite Programmi
 ng Languages:\n\nFortran\, C or C++. It is not possible to do the exercise
 s in Java.\n\nPre-requisite setup\n\nAttendees should bring their own lapt
 op (Windows/Mac/Linux) and will need the required software installed:\n\n
  \n\nTimetable\n\nDay 1  Thursday 14th May\n\n09:30  Message-Passing Co
 ncepts\n10:15  Practical: Parallel Traffic Modelling\n11:00  Break\n11:3
 0  MPI Programs\n12:00  MPI on ARCHER\n12:15  Practical: Hello World\n1
 3:00  Lunch\n14:00  Point-to-Point Communication\n14:30  Practical: Pi\
 n15:30  Break\n16:00  Communicators\, Tags and Modes\n16:45 Practical: P
 ing-Pong\n17:30  Finish\n\nDay 2  Friday 15th May\n\n09:30  Pi Solution
 \n10:00  Non-Blocking Communication\n10:30  Practical: Message Round a R
 ing\n11:00  Break\n11:30  Collective Communicaton\n12:00  Practical: Co
 llective Communication\n13:00  Lunch\n14:00  Virtual Topologies\n14:30 
  Practical: Message Round a Ring (cont.)\n15:30  Break\n16:00  Derived D
 ata Types\n16:45  Practical: Message Round a Ring (cont.)\n17:30  Finish
 \n\nDay 3  Friday 22nd May\n\n09:30  MPI Quiz\n10:15  Introduction to t
 he Case Study\n10:30  Practical: Case Study\n11:00  Break\n11:30  Pract
 ical: Case Study (cont.)\n13:00  Lunch\n14:00  Designing MPI Programs\n1
 5:00 Individual Consultancy Session\n15:30  Finish\nhttps://events.prace-
 ri.eu/event/1017/
SUMMARY:[ONLINE] Message-passing programming with MPI @ EPCC
URL;VALUE=URI:https://events.prace-ri.eu/event/1017/
END:VEVENT
END:VCALENDAR