Computing boot camp for UD scientists
June 21, 2021
Learn how to program GPUs
Does your scientific code take hours to days to run? Would you like to learn how to improve your code and run on large-scale high-performance computing systems?
On July 8–9, the University of Delaware HPC Bootcamp will teach you how to write a portable parallel program that can run on multicore CPUs and accelerators like GPUs.
NVIDIA and OpenACC are hosting this exclusive GPU boot camp for UD scientists. It is open to undergraduate and graduate students, faculty, staff, postdoctoral researchers, and anyone else who is interested. The boot camp will be held over two days, four hours each day.
The digital boot camp uses a directive-based programming model, OpenACC, that can be used to migrate scientific code from traditional CPUs to Graphic Processor Units (GPUs). The OpenACC ported code also will run on the CPUs, thus creating a performant portable code that is running across the CPUs and the GPUs.
OpenACC is being widely used for porting many applications including those from solar/nuclear/biophysics, medicine, agriculture, climate modeling, molecular dynamics and so on. Recent success stories at UD include work by Prof. Juan Perilla and team in the Department of Chemistry and Biochemistry and Prof. Sunita Chandrasekaran in the Department of Computer and Information Sciences.
A mini-application challenge will feature two prizes:
- First place team will receive a $500 cash prize and a certificate sponsored by UD's Data Science Institute (DSI); and
- Second place team will receive a complimentary copy of OpenACC for Programmers: Concepts and Strategies, written by Sunita Chandrasekaran of UD and Guido Juckeland from Helmholtz-Zentrum Dresden-Rossendorf.
Register by June 23 at this webpage.
Questions? Contact Sunita Chandrasekaran at firstname.lastname@example.org.