Executed and finished Sprints
Already accepted sprints can be found here. For more information beyond the PDF, please check out our Gitlab Wiki and our Lessons-learned page.
Nr |
Model |
Service Description
|
Documentation
|
|
01 |
Analysis of ART code for GPU porting |
||
|
02 |
Online coupling mHM into ICON using YAC |
||
|
03 |
Port FESOM 2.1 to JUWELS booster and Levante- GPU |
||
|
04 |
Port ParFlow to AMD GPUs, Inspection of RAPID Memory Manager and Hipification, Performance Analysis |
||
|
05 |
Optimise the data transfers between host (CPU) and device (GPU) |
||
|
06 |
Updating the remaining non-lazy preprocessor functions to be memory efficient and updating ESMValCore |
||
|
07 |
Concurrent HAMOCC on GPU |
||
|
08 |
Couple MESSy to ICON via the ICON Community Interface (ComIn) |
||
|
09 |
Develop a concept and provide a first implementation of a framework for Lagrangian transport modeling |
||
|
10 |
QUINCY in ICON-Land |
Stepwise (mode based) port of the IQ code to GPUs based on the established workflow followed for the ICON GPU implementation with OpenACC |
|
|
12 |
Porting those of the PALM modules related to urban processes (especially radiation) to GPUs |
||
|
14 |
Coupling CLEO to ICON with YAC |
Technical Information
The requirements to apply for a sprint are summarized in a here. Useful insights, e.g., lessons learned as well as FAQ are continuously added.
Slides, technical exercises and data examples from our workshops are listed on the workshop sites.