BEGIN:VCALENDAR VERSION:2.0 PRODID:icalendar-ruby CALSCALE:GREGORIAN BEGIN:VEVENT DTSTAMP:20260622T105758Z UID:b4be6cf0-6e51-44cd-9ba5-7201da2771f2 DTSTART:20130703T080000Z DTEND:20130705T163000Z DESCRIPTION:\n Course Description\n\n The world’s largest supercomputers are used almost exclusively to run applications which are parallelised usi ng Message Passing. The course covers all the basic knowledge required to write parallel programs using this programming model\, and is directly app licable to almost every parallel computer architecture.\n\n Parallel progr amming by definition involves co-operation between processes to solve a co mmon task. The programmer has 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 separat e processes that communicate and synchronise by explicitly sending each ot her messages. All these parallel operations are performed via calls to som e message-passing interface that is entirely responsible for interfacing w ith the physical communication network linking the actual processors toget her. This course uses the de facto standard for message passing\, the Mess age Passing Interface (MPI). It covers point-to-point communication\, non- blocking operations\, derived datatypes\, virtual topologies\, collective communication and general design issues.\n\n The course is normally delive red in an intensive three-day format using EPCC’s dedicated training fac ilities. It is taught using a variety of methods including formal lectures \, practical exercises\, programming examples and informal tutorial discus sions. This enables lecture material to be supported by the tutored practi cal sessions in order to reinforce the key concepts.\n\n Intended Learning Outcomes\n\n On completion of this course students should be able to:\n\n Understand the message-passing model in detail.\n \n Implement standard message-passing algorithms in MPI.\n \n Debug simple MPI codes.\n \n Me asure and comment on the performance of MPI codes.\n \n Design and implem ent efficient parallel programs to solve regular-grid problems.\n\n Pre-re quisite Programming Languages:\n\n Fortran\, C or C++. It is not possible to do the exercises in Java.\n\n Draft timetable\n\n Day 1\n\n 09:30 Mess age-Passing Concepts\n \n 10:15 Practical: Parallel Traffic Modelling\n \ n 11:00 COFFEE\n \n 11:30 MPI Programs\n \n 12:00 MPI on Ness and HECTo R\n \n 12:15 Practical: Hello World\n \n 13:00 LUNCH\n \n 14:00 Point-t o-Point Communication\n \n 14:30 Practical: Pi\n \n 15:30 TEA\n \n 16:0 0 Communicators\, Tags and Modes\n \n 16:45 Practical: Ping-Pong\n \n 17 :30 CLOSE\n\n Day 2\n\n 09:30 Non-Blocking Communication\n \n 10:00 Prac tical: Message Round a Ring\n \n 11:00 COFFEE\n \n 11:30 Collective Comm unicaton\n \n 12:00 Practical: Collective Communication\n \n 13:00 LUNCH \n \n 14:00 Virtual Topologies\n \n 14:30 Practical: Message Round a Rin g (cont.)\n \n 15:30 TEA\n \n 16:00 Derived Data Types\n \n 16:45 Pract ical: Message Round a Ring (cont.)\n \n 17:30 CLOSE\n\n Day 3\n\n 09:30 Introduction to the Case Study\n \n 10:00 Practical: Case Study\n \n 11: 00 COFFEE\n \n 11:30 Practical: Case Study (cont.)\n \n 12:15 Designing MPI Programs\n \n 13:00 LUNCH\n \n 14:00 Individual Consultancy Session\ n \n 15:30 CLOSE\n\n\nhttps://events.prace-ri.eu/event/158/ SUMMARY:Message-Passing Programming with MPI @ EPCC URL;VALUE=URI:https://events.prace-ri.eu/event/158/ END:VEVENT END:VCALENDAR