As far as I know, you don’t have to use the MPI_ASYNC_PROGRESS settings with Intel MPI, either. However, you may see a huge boost in performance, at least for the onesided operations using MPI Windows.
Bruce Palmer
From: viet via discuss <discuss@mpich.org>
Sent: Thursday, January 7, 2021 4:08 AM
To: Zhou, Hui <zhouh@anl.gov>; discuss@mpich.org
Cc: teivml@gmail.com
Subject: Re: [mpich-discuss] Question about non-blocking MPI - 3
Check twice before you click! This email originated from outside PNNL.
Dear Hui Zhou,
Thank you for your response to this thread.
Could you give me some references on this issue?
It seems that the below environment setting is necessary.
MPICH_ASYNC_PROGRESS=1
Thank you for anything you can provide.
Best, Viet.
On Tue, Jan 5, 2021 at 5:59 AM Zhou, Hui <zhouh@anl.gov> wrote:
You don’t need any special settings to use non-blocking MPI collectives in MPICH.
--
Hui Zhou
From: viet via discuss <discuss@mpich.org>
Date: Friday, December 4, 2020 at 7:12 AM
To: discuss@mpich.org <discuss@mpich.org>
Cc: teivml@gmail.com <teivml@gmail.com>
Subject: [mpich-discuss] Question about non-blocking MPI - 3Hello everyone,
What is the environment setting for the use of non-blocking MPI-3 * collectives in mpich3.3?
In Intel MPI, the setting is as follows
$ export I_MPI_ASYNC_PROGRESS=1
$ export I_MPI_ASYNC_PROGRESS_PIN=<CPU list>
Is there an equivalent setting in mpich3.3?
Thank you,
Viet.