Skip to content

To port MeshMaterialSynchronizer::synchronizeMaterialsInCells() with the accelerator API #1168

Description

@DavidDureau

Is it possible to port the synchronizeMaterialsInCells() with the accelerator API?

Indeed, when a code is running with AcceleratorRuntime=cuda, then the main part of synchronizeMaterialsInCells() is running on CPU and the mat_presence.synchronize() is "running" on GPU, which implies H<->D transfers.

We have ported the same kind of algorithm in a code so we have an example.

Metadata

Metadata

Assignees

No one assigned

    Labels

    arcaneArcane Componentarcane:acceleratorArcane Accelerator API/Runtimearcane:materialsManagement of Constituents (Material/Environment)enhancementNew feature or request

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions