Multi-GPU Acceleration of Interface Problems in FEniCS, with Application to Cardiac Modeling (Q7414766)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

No description defined
Language Label Description Also known as
default for all languages
No label defined
    English
    Multi-GPU Acceleration of Interface Problems in FEniCS, with Application to Cardiac Modeling
    No description defined

      Statements

      Multi-GPU Acceleration of Interface Problems in FEniCS, with Application to Cardiac Modeling (English)
      0 references
      2 November 2026
      0 references
      math.NA
      0 references
      cs.DC
      0 references
      cs.MS
      0 references
      Edoardo Centofanti
      0 references
      Benjamin Pachev
      0 references
      Joshua Steyer
      0 references
      James D. Trotter
      0 references
      Pietro Benedusi
      0 references
      We extend cuDOLFINx, a GPU-enabled extension of FEniCS, to support mixed-dimensional interface problems on GPUs while preserving the high-level FEniCS programming model, enabling end-to-end GPU execution of an elliptic interface problem from cardiac modeling. This results in strong and weak scaling advantages over CPU-based implementations for certain problem sizes and geometries.
      0 references
      GPU Acceleration
      0 references
      Finite Element Methods
      0 references
      Interface Problems
      0 references
      Cardiac Modeling
      0 references

      Identifiers

      0 references