Back To Schedule
Thursday, August 9 • 8:30am - 10:00am
Achieving High Performance Python in Modern HPC Workflow

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
The evolution in usage of popular scripting languages such as Python has led to the language's growing popularity in HPC work. With more and more numerical workloads being written in Python, getting the most out of one's mixed C and Python code becomes challenging with the intricacies and restrictions normally present in the Python language. However, recent advancements in Python tools has allowed for additional options for mixed codebases–with the proper knowledge of how to design, measure, and publish such Python packages in a compute center, a high-productivity Python compute environment can easily coexist with existing HPC code and workflows.
 This presentation will cover the theory behind Python HPC performance, the tools used to convert portions of the code back and forth between Python and C code, determining and profiling the performance of a mixed application, and finally building and deploying applications within a compute center. Examples will be shown of each of the critical methods above will provide reinforcement of the concepts, followed afterwards with a Q&A session.


David Liu

David is a Technical Consultant Engineer at Intel Corporation in Austin, Texas, where he represents Intel's Python products and projects. He is focused on solving customer problems in Python while simultaneously developing and shaping Intel's software products to match customer needs... Read More →

Thursday August 9, 2018 8:30am - 10:00am MDT
Room 207

Attendees (4)