You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Large queues (> 2GB) are broken into smaller pieces (tagged with tags::piece) to work around MPI limitations. detail::VectorWindow<char> is used to put subsets of the MemoryBuffers into MPI i[s]send routines without having to make a copy. shared_ptrs are used to keep track of the original buffer as long as some in-flight message needs it.
We could use the same mechanism during, e.g., all-to-all operations to avoid making copies between buffers.
The text was updated successfully, but these errors were encountered:
Large queues (
> 2GB
) are broken into smaller pieces (tagged withtags::piece
) to work around MPI limitations.detail::VectorWindow<char>
is used to put subsets of theMemoryBuffer
s into MPIi[s]send
routines without having to make a copy.shared_ptr
s are used to keep track of the original buffer as long as some in-flight message needs it.We could use the same mechanism during, e.g.,
all-to-all
operations to avoid making copies between buffers.The text was updated successfully, but these errors were encountered: