2020-01-22 Introduction to GPU programming (09:00-17:00)

GPU Computing

Graphical Processor Units (GPUs) can be used to speed up certain scientific computing tasks by orders of magnitude. Thus GPUs are now available on many HPC computing clusters including the university Linux cluster UBELIX. 

Goal 

Learn about computing methods for GPUs and run your first CUDA code.

Learning outcomes - participants know/can

• About GPU computingand trends in HPC
• GPU hardware architecture
• How to program on GPU
• CUDA syntax

Target group

• UNIBE staff members, students and potential users involved in scientific research

Prerequisites 

• Participants must bring own laptops and should have a minimal Linux and C/C++ programming knowledge 

Methods

• The training alternates between short theoretical introductions and programming exercises.

Certificate 

• A certificate will be delivered to participants who have attended the whole training. The certificate is valid as two elective half days for the CAS Applied Data Science.

Coaches

• The coaches are local or external experts

Time : 2020-01-22 09:00 - 17:00
Location :  University of Bern, Room 120, Mittelstrasse 43, University of Bern

Training language: English
Participants : Max 20
Registraion : Mandatory
Coaches : Dr. Simon Grimm (lecturer and coach), Dr. Alexander Kashev (tutor)
Prerequisites : Laptop, Linux and C/C++ programming experience 
Certificate : Certificate for full training attendance 

The Science IT Support is there to boost your research by supporting you solving computing challenges. 
Your code doesn't compile, you need more computing power, more storage, a data mangament plan and
so on -  scits.unibe.ch.