Challenges of Performance Portability for Fortran Unstructured Mesh Codes Your name Objectives: o Investigates how different approaches to parallel optimization impact the performance portability o Explores the productivity challenges Limitation of software tool and compiler support Accomplishments: o Develop initial ports for OpenMP CPU, OpenMP offload GPU and CUDA o Compare the performance across the Exascale Impact: o Face some difficulties towards these approaches o Encourage users and venders to focus on the productive pathways to developing Fortran applications o Draw attention for support on Fortran tools Abigail Hsu Stony Brook University [email protected]
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Challenges of Performance Portability for Fortran Unstructured Mesh Codes
Your name
Objectives:
o Investigates how different approaches to parallel optimization impact the performance portability
o Explores the productivity challenges Limitation of software tool and compiler support
Accomplishments:
o Develop initial ports for OpenMP CPU, OpenMP offload GPU and CUDA
o Compare the performance across the Exascale Impact:
o Face some difficulties towards these approaches o Encourage users and venders to focus on the
productive pathways to developing Fortran applications o Draw attention for support on Fortran tools
Abigail Hsu Stony Brook University [email protected]
Related Documents
