Back to Eqiora docs Skip to main content

MpiCudaAdmittedExecutionAdapter

eqiora::backends::mpi_cuda

Trait MpiCudaAdmittedExecutionAdapter 

pub trait MpiCudaAdmittedExecutionAdapter {
    // Required method
    fn execute_admitted_mpi_cuda(
        &mut self,
        admitted: AdmittedExecution<'_>,
    ) -> Result<AcceptedMpiCudaLinearExecution, Diagnostic>;
}
Description

L3 seam consuming one exact distributed-CUDA L2 admission.

Required Methods§

fn execute_admitted_mpi_cuda( &mut self, admitted: AdmittedExecution<'_>, ) -> Result<AcceptedMpiCudaLinearExecution, Diagnostic>

Signature links:execute_admitted_mpi_cudaResultAcceptedMpiCudaLinearExecutionDiagnostic

Execute host-staged MPI Krylov with one resident CUDA CSR action on every rank.

§Errors

Returns a common diagnostic for binding/runtime/device topology, resident setup, local action, MPI protocol, receipt, or summary drift.

Implementors§