Reference
MPILargeCounts.MPILargeCounts — Module
Chunked communication based on MPI.jl with arbitrary-size data.
MPILargeCounts.Irecv! — Method
req = Irecv!(recvbuf, comm::Comm;
source::Integer=MPI.ANY_SOURCE, tag::Integer=MPI.ANY_TAG)Starts a nonblocking receive into the buffer recvbuf from MPI rank source of communicator comm using the message tag tag.
MPILargeCounts.Isend — Method
Isend(data, comm::Comm; dest::Integer, tag::Integer=0)Starts a nonblocking send of data to MPI rank dest of communicator comm using with the message tag tag.
MPILargeCounts.Recv! — Function
data = Recv!(recvbuf, comm::Comm;
source::Integer=MPI.ANY_SOURCE, tag::Integer=MPI.ANY_TAG)Completes a blocking receive into the buffer recvbuf from MPI rank source of communicator comm using with the message tag tag.
MPILargeCounts.Send — Method
Send(buf, comm::Comm; dest::Integer, tag::Integer=0)Perform a blocking send from the buffer buf to MPI rank dest of communicator comm using the message tag tag.
Send(obj, comm::Comm; dest::Integer, tag::Integer=0)Complete a blocking send of an isbits object obj to MPI rank dest of communicator comm using with the message tag tag.
MPILargeCounts.allreduce — Method
allreduce(obj, op, comm::Comm; root::Integer=0)
allreduce!(obj, op, comm::Comm; root::Integer=0)Performs elementwise reduction using the operator op on the object obj across all ranks in comm, returning the reduced object on all ranks.
MPILargeCounts.bcast — Method
bcast(obj, comm::Comm; root::Integer=0)Broadcast the object obj from rank root to all processes on comm. This is able to handle arbitrary data.
MPILargeCounts.isend — Method
isend(obj, comm::Comm; dest::Integer, tag::Integer=0)Starts a nonblocking send of using a serialized version of obj to MPI rank dest of communicator comm using with the message tag tag.
Returns the communication Request for the nonblocking send.
MPILargeCounts.mpi_execute_on_root — Method
mpi_execute_on_root(F::A, args...; kwargs...)Execute the function F with arguments args... and keyword arguments kwargs... on the root of the communicator comm (default: MPI.COMM_WORLD). If blocking is set to true (default: false), a barrier is performed before and after the execution to synchronize all ranks. Returns the result of F on the root, and nothing on all other ranks.
MPILargeCounts.mpi_execute_on_root_and_bcast — Method
mpi_execute_on_root_and_bcast(F::A, args...; kwargs...)Execute the function F with arguments args... and keyword arguments kwargs... on the root of the communicator comm (default: MPI.COMM_WORLD), and broadcast the result to all ranks. Broadcasts the result of F on all ranks.
MPILargeCounts.recv — Function
obj = recv(comm::Comm;
source::Integer=MPI.ANY_SOURCE, tag::Integer=MPI.ANY_TAG)Completes a blocking receive of a serialized object from MPI rank source of communicator comm using with the message tag tag.
MPILargeCounts.reduce — Method
reduce(obj, op, comm::Comm; root::Integer=0)
reduce!(obj, op, comm::Comm; root::Integer=0)Performs elementwise reduction using the operator op on the object obj on the root rank root of comm, returning the reduced object on rank root.
MPILargeCounts.send — Method
send(obj, comm::Comm; dest::Integer, tag::Integer=0)Complete a blocking send using a serialized version of obj to MPI rank dest of communicator comm using with the message tag tag.
MPILargeCounts.split_buffer — Method
split_buffer(vec)Split an AbstractVector into a Vector{MPI.Buffer} views without copying.