Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

docs: Improve documentation for collective operations #1084

Closed
wants to merge 4 commits into from
Closed
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 12 additions & 7 deletions heat/core/communication.py
Original file line number Diff line number Diff line change
Expand Up @@ -1371,8 +1371,9 @@ def Alltoall(
recv_axis: int = None,
):
"""
All processes send data to all processes: The jth block sent from process i is received by process j and is
placed in the ith block of recvbuf.
Uniform data exchange between all processes: Collective communication operation that allows equal amount of data exchange
with all other processes. The jth block sent from process i is received by process j and is placed in the ith block of
recvbuf.

Parameters
----------
Expand Down Expand Up @@ -1407,8 +1408,9 @@ def Alltoallv(
recv_axis: int = None,
):
"""
v-call of Alltoall: All processes send different amount of data to, and receive different amount of data
from, all processes
Non-uniform data exchange between all processes: Generalized collective operation in which all processes exchange variable
amounts of data from all other processes. The j-th block of process i's local sendbuf sent is stored in the ith block of
process j's local recvbuf.

Parameters
----------
Expand Down Expand Up @@ -1443,7 +1445,9 @@ def Ialltoall(
recv_axis: int = None,
) -> MPIRequest:
"""
Nonblocking Alltoall
Nonblocking uniform data exchange between all processes: Collective communication operation that allows equal amount of data
exchange with all other processes. The jth block sent from process i is received by process j and is placed in the ith block
of recvbuf.

Parameters
----------
Expand Down Expand Up @@ -1473,8 +1477,9 @@ def Ialltoallv(
recv_axis: int = None,
) -> MPIRequest:
"""
Nonblocking v-call of Alltoall: All processes send different amount of data to, and receive different amount of
data from, all processes
Nonblocking variable data exchange between all processes: Generalized collective operation in which all processes exchange
variable amounts of data from all other processes. The j-th block of process i's local sendbuf sent is stored in the ith block
of process j's local recvbuf.

Parameters
----------
Expand Down