Purdue hosting Intel parallel programming, acceleration workshop

October 6, 2015  9:00am – October 7, 2015  4:00pm
Stewart Center, Room 202

Purdue will host a two-day parallel programming workshop by Intel in October designed to assist researchers and software developers with writing and optimizing code to take full advantage of hardware like Intel’s Xeon processor and Xeon Phi coprocessor, both features of Purdue’s Conte cluster research supercomputer.

The two-day training session is for Purdue and non-Purdue faculty, staff and students, as well as area professional software engineers and architects. “Parallel Programming and Optimization with Intel Xeon Phi Coprocessors” includes an introductory seminar from 9 a.m. to 4 p.m. Tuesday, Oct. 6, and a hands-on lab session from 9 a.m. to 4 p.m. Wednesday, Oct 7, in the {Stewart Center}, Room 202. Attendance at the introductory seminar is a prerequisite for the lab session.

Those interested can get more information and register for the introductory seminar here and the lab session here. There is no cost to register. Intel and ITaP are sponsoring the event.

Check-in begins at 8:30 a.m. each day. Lunch will be provided both days. Participants need to bring a laptop with Windows (XP or higher), Mac OS X (version 10.5 or later) or Linux installed.

The training session will cover such topics as:

  • An overview of programming models and parallel frameworks, such as MPI and OpenMP, for multi-core processors and many-core coprocessors.

  • Methods and tools for finding bottlenecks and optimizing code performance, such as scalar math, vectorization and multithreading.

  • Real-life examples of code and optimization techniques.

Staff from ITaP Research Computing will be available to interact with people attending the session, says Stephen Harrell, an ITaP senior scientific application analyst and training coordinator.

ITaP Research computing operates Conte and Purdue’s other Community Cluster Program supercomputers for faculty researchers and their students, including the new Rice cluster, which made the latest list of the world’s TOP500 supercomputers, and the new Snyder cluster for memory-intensive life sciences applications. Conte also is a TOP500 supercomputer.

For more information, email rcac-help@purdue.edu.

Originally posted: September 17, 2015  11:28am