BEGIN:VCALENDAR VERSION:2.0 PRODID:icalendar-ruby CALSCALE:GREGORIAN BEGIN:VEVENT DTSTAMP:20260706T215040Z UID:9e24c773-8f9d-47d3-b750-246f69b425ec DTSTART:20181015T080000Z DTEND:20181019T150000Z DESCRIPTION:The registration to this course is now open. Please\, bring yo ur own laptop. All the PATC courses at BSC are free of charge.\n\nCourse C onvener: Xavier Martorell\n\nLOCATION: UPC Campus Nord premises.Vertex Bui lding\, Room VS208\n\nLevel: \n\nIntermediate: For trainees with some the oretical and practical knowledge\, some programming experience.\n\nAdvance d: For trainees able to work independently and requiring guidance for solv ing complex problems.\n\nAttendants can bring their own applications and w ork with them during the course for parallelization and analysis.\n\nPrere quisites: Fortran\, C or C++ programming. All examples in the course will be done in C\n\nObjectives: \n\nThe objectives of this course are to und erstand the fundamental concepts supporting message-passing and shared mem ory programming models. The course covers the two widely used programming models: MPI for the distributed-memory environments\, and OpenMP for the s hared-memory architectures. The course also presents the main tools develo ped at BSC to get information and analyze the execution of parallel applic ations\, Paraver and Extrae. It also presents the Parallware Assistant too l\, which is able to automatically parallelize a large number of program s tructures\, and provide hints to the programmer with respect to how to cha nge the code to improve parallelization. It deals with debugging alternati ves\, including the use of GDB and Totalview.\n\nThe use of OpenMP in conj unction with MPI to better exploit the shared-memory capabilities of curre nt compute nodes in clustered architectures is also considered. Paraver wi ll be used along the course as the tool to understand the behavior and per formance of parallelized codes. The course is taught using formal lectures and practical/programming sessions to reinforce the key concepts and set up the compilation/execution environment.\n\nAttendants can bring their ow n applications and work with them during the course for parallelization an d analysis.\n\nLearning Outcomes:\n\nThe students who finish this course w ill be able to develop benchmarks and applications with the MPI\, OpenMP a nd mixed MPI/OpenMP programming models\, as well as analyze their executio n and tune their behaviour in parallel architectures.\n\nAgenda:\n\nDay 1 (Monday)\n\nSession 1 / 9:30 am – 1:00 pm (2 h lectures\, 1 h practical) \n\n1. Introduction to parallel architectures\, algorithms design and perf ormance parameters\n\n2. Introduction to the MPI programming model\n\n3. P ractical: How to compile and run MPI applications\n\n \n\nSession 2 / 2:0 0pm – 5:30 pm (2h lectures\, 1h practical)\n\n1. MPI: Point-to-point com munication\, collective communication\n\n2. Practical: Simple matrix compu tations\n\n3. MPI: Blocking and non-blocking communications\n\n \n\n Day 2 (Tuesday)\n\nSession 1 / 9:30 am - 1:00 pm (1.5 h lectures\, 1.5 h prac tical)\n\n1. MPI: Collectives\, Communicators\, Topologies\n\n2. Practical : Heat equation example\n\n \n\nSession 2 / 2:00 pm - 5:30 pm (1.5 h lect ures\, 1.5 h practical)\n\n1. Introduction to Paraver: tool to analyze and understand performance\n\n2. Practical: Trace generation and trace analys is\n\n \n\nDay 3 (Wednesday)\n\nSession 1 / 9:30 am - 1:00 pm (1 h lectur e\, 2h practical)\n\n1. Parallel debugging in MareNostrumIII\, options fro m print to Totalview\n\n2. Practical: GDB and IDB\n\n3. Practical: Totalvi ew\n\n4. Practical: Valgrind for memory leaks\n\n \n\nSession 2 / 2:00 pm - 5:30 pm (2 h lectures\, 1 h practical)\n\n1. Shared-memory programming models\, OpenMP fundamentals\n\n2. Parallel regions and work sharing const ructs\n\n3. Synchronization mechanisms in OpenMP\n\n4. Practical: heat dif fusion in OpenMP\n\n \n\nDay 4 (Thursday)\n\nSession 1 / 9:30am – 1:00 pm  (2 h practical\, 1 h lectures)\n\n1. Tasking in OpenMP 3.0/4.0/4.5\n\ n2. Programming using a hybrid MPI/OpenMP approach\n\n3. Practical: multis ort in OpenMP and hybrid MPI/OpenMP\n\n \n\nSession 2 / 2:00pm – 5:30 p m (1.5 h lectures\, 1.5 h practical)\n\n1. Parallware: guided parallelizat ion\n\n2. Practical session with Parallware examples\n\n \n\nDay 5 (Frida y)\n\nSession 1 / 9:30 am – 1:00 pm (2 hour lectures\, 1 h practical)\n\ n1. Introduction to the OmpSs programming model\n\n2. Practical: heat equa tion example and divide-and-conquer\n\n \n\nSession 2 / 2:00pm – 5:30 p m (1 h lectures\, 2 h practical)\n\n1. Programming using a hybrid MPI/OmpS s approach\n\n2. Practical: heat equation example and divide-and-conquer\n \n \n\nEND of COURSE\n\nhttps://events.prace-ri.eu/event/756/ SUMMARY:Parallel Programming Workshop @ BSC URL;VALUE=URI:https://events.prace-ri.eu/event/756/ END:VEVENT END:VCALENDAR