Difference between revisions of "Optimization of a lagrangian cloud parcel model for use in the global climate model ECHAM6.1-HAM2.2"

From SNIC Documentation
Jump to: navigation, search
(Created page with "{{project info |description=Serial optimization of a cloud parcel model for embedding in a the ECHAM-HAM global climate model. |fields=Climate research |financing=SNIC |active=No...")
 
 
Line 8: Line 8:
 
}}
 
}}
  
The project aims to optimize the runtime performance of a largangian cloud parcel model. It is planned for this code to be embedded in the ECHAM6.1-HAM2.2 in a novel approach of modeling cloud microphysics within a global climate model.
+
The [[Media:Dan_Partridge_proposal.pdf|project]] aims to optimize the runtime performance of a largangian cloud parcel model. It is planned for this code to be embedded in the ECHAM6.1-HAM2.2 in a novel approach of modeling cloud microphysics within a global climate model.
  
 
Project objectives:
 
Project objectives:

Latest revision as of 09:37, 13 October 2015

Name Optimization of a lagrangian cloud parcel model for use in the global climate model ECHAM6.1-HAM2.2
Description Serial optimization of a cloud parcel model for embedding in a the ECHAM-HAM global climate model.
Project financing   SNIC
Is active No
Start date 2015-04-01
End date 2015-05-31

The project aims to optimize the runtime performance of a largangian cloud parcel model. It is planned for this code to be embedded in the ECHAM6.1-HAM2.2 in a novel approach of modeling cloud microphysics within a global climate model.

Project objectives:

  • Complete detailed profiling of the cloud parcel model code
  • Analyze the cloud parcel model code to identify code sections and routines to be further evaluated for performance optimization.
  • Propose a set of recommendations for code changes based on the performance analysis.
  • Make code changes identified to be very fast to implement based on the profiling results.
  • Evaluate whether openmp parallelisation should be feasible and how much work that would entail.

Members

 CentreRoleField
Chandan Basu (NSC)NSCApplication expertComputational science