BEGIN:VCALENDAR VERSION:2.0 PRODID:icalendar-ruby CALSCALE:GREGORIAN BEGIN:VEVENT DTSTAMP:20260704T193443Z UID:8bf5a62a-740a-4397-b009-335efdc69a03 DTSTART:20170626T060000Z DTEND:20170630T160000Z DESCRIPTION:The Barcelona Supercomputing Center (BSC) in association with U niversitat Politecnica de Catalunya (UPC) has been awarded by NVIDIA as a GPU Center of Excellence. BSC and UPC currently offer a number of courses covering CUDA architecture and programming languages for parallel computin g. Please contact us for possible collaborations.\n\nThe 8th edition of th e Programming and Tuning Massively Parallel Systems summer school (PUMPS) is aimed at enriching the skills of researchers\, graduate students and te achers with cutting-edge technique and hands-on experience in developing a pplications for many-core processors with massively parallel computing res ources like GPU accelerators.\n\n\n Summer School Co-Directors: Mateo Vale ro (BSC and UPC) and Wen-mei Hwu (University of Illinois at Urbana-Champai gn)\n \n \n Local Organizers: Antonio J. Peña (BSC)\, Victor Garcia (BSC and UPC)\, and Pau Farre (BSC)\n \n \n Dates:\n\n \n Applications due: Ap ril 30\, 2017\n\n \n Due to space limitations\, early application is st rongly recommended. You may also be suggested to attend an online prerequi site training on basic CUDA programming before joining PUMPS.\n \n \n \n Notification of acceptance: May 15\, 2017\n \n \n Summer school: Ju ne 26-30\, 2017\n \n \n\n Organized by:\n\n \n Barcelona Supercomputing Center (BSC)\n \n \n University of Illinois at Urbana-Champaign (Univer sity of Illinois)\n \n \n Universitat Politecnica de Catalunya (UPC)\n \n \n HiPEAC Network of Excellence (HiPEAC)\n \n \n PUMPS is part of this year PRACE Advanced Training Centre program\n \n \n\n The following is a list of some of the topics that will be covered during the course:\n \n \n CUDA Algorithmic Optimization Strategies\n \n \n Dealing with Sp arse and Dynamic data\n \n \n Efficiency in Large Data Traversal\n \n \n Reducing Output Interference\n \n \n Controlling Load Imbalance an d Divergence\n \n \n Acceleration of Collective Operations\n \n \n D ynamic Parallelism and HyperQ\n \n \n Debugging and Profiling CUDA Code \n \n \n Multi-GPU Execution\n \n \n Architecture Trends and Implica tions\n \n \n Introduction to OmpSs and to the Paraver analysis tool\n \n \n OmpSs: Leveraging GPU/CUDA Programming\n \n \n Hands-on Labs: CUDA Optimizations on Scientific Codes\; OmpSs Programming and Tuning\n \ n \n\n Instructors:\n\n \n Distinguished Lecturers: Wen-mei Hwu (Universi ty of Illinois at Urbana-Champaign) and David Kirk (NVIDIA Corporation )\n \n \n Invited Lecturer: Juan Gómez-Luna (Universidad de Córdoba)\n \n \n BSC / UPC Lecturers: Xavier Martorell and Xavier Teruel\n \n \n Teaching Assistants: Abdul Dakkak\, Carl Pearson\, Simon Garcia de Gonzal o\, Marc Jorda\, Pau Farre\, Javier Bueno\, Aimar Rodriguez\n \n \n\n Pre requisites for the course are:\n\n \n Basic CUDA knowledge is required to attend the course. Applicants that cannot certify their experience in CUD A programming will be asked to take a short on-line course covering the ne cessary introductory topics\n \n \n C\, C++\, Java\, or equivalent prog ramming knowledge. Skills in parallel programming will be helpful\n \n \n Preliminary Overview\n\n\n Registration for the course is free. We expect our sponsors will cover academic applicants' marginal expenses such as mea ls. Please note that travel and lodging are not covered. Applicants from n on-academic institutions (companies)\, please contact us at bcw2017@bcw.ac .upc.edu for sponsorship possibilities.\n \n\n By the end of the summer sc hool\, participants will:\n\n \n Be able to design algorithms that are su itable for accelerators.\n \n \n Understand the most important architec tural performance considerations for developing parallel applications.\n \n \n Be exposed to computational thinking skills for accelerating appli cations in science and engineering.\n \n \n Engage computing accelerato rs on science and engineering breakthroughs.\n \n \n\n Programming Langua ges: CUDA\, MPI\, OmpSs\, OpenCL\n \n \n Hands-on Labs: Afternoon labs wit h teaching assistants for each audience/level.\n\n \n Participants are ex pected to bring their own laptops to access the servers with GPU accelerat ors.\n \n \n The afternoon lab sessions will provide hands-on experienc e with various languages and tools covered in the lectures and will compri se a brief introduction to the programming assignments\, followed by indep endent work periods. Teaching assistants will be available in person and o n the web to help with assignments.\n \n \n\n\nhttps://events.prace-ri.eu /event/610/ SUMMARY:8th Programming and Tuning Massively Parallel Systems summer school (PUMPS)@BSC - UPC URL;VALUE=URI:https://events.prace-ri.eu/event/610/ END:VEVENT END:VCALENDAR