[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Octave-bug-tracker] [bug #60251] [octave-forge] (parallel) strange exec
|
From: |
Olaf Till |
|
Subject: |
[Octave-bug-tracker] [bug #60251] [octave-forge] (parallel) strange execution times |
|
Date: |
Wed, 17 Mar 2021 14:20:18 -0400 (EDT) |
|
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Firefox/78.0 |
Follow-up Comment #1, bug #60251 (project octave):
Starting with 4.0.0, parallel uses freshly started Octave sessions as parallel
workers. These are started by the first call to parcellfun. (More are started
if 'parcellfun's first argument is larger than in any call before.) This makes
the first call take longer time. 9 seconds for 'nproc' starts of 'octave-cli'
seems long, though...
That the second call takes longer than in version 3.1.3 may be expected
(although a comparison with different Octave compilations is only rough),
since the communication between the master and slave sessions in versions
starting with 4.0.0 constitutes some overhead.
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?60251>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/