climateprediction.net (CPDN) home page
Thread 'I feel sorry for this machine...'

Thread 'I feel sorry for this machine...'

Message boards : Number crunching : I feel sorry for this machine...
Message board moderation

To post messages, you must log in.

AuthorMessage
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1120
Credit: 17,202,915
RAC: 2,154
Message 68267 - Posted: 11 Feb 2023, 19:55:47 UTC

This poor machine gets suspended about 14 times per minute. No wonder it progresses so slowly. It is a wonder it gets anywhere at all.

Task 22306830
Name 	oifs_43r3_ps_0561_2021050100_123_990_12206681_0
Workunit 	12206681

<core_client_version>7.18.1</core_client_version>
<![CDATA[
<message>
process exited with code 1 (0x1, -255)</message>
<stderr_txt>

  22:59:51 STEP  819 H= 819:00 +CPU= 42.573
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
Suspend request received from the BOINC client, suspending the child process
Resuming the child process
  23:00:47 STEP  820 H= 820:00 +CPU= 27.309

ID: 68267 · Report as offensive     Reply Quote
Glenn Carver

Send message
Joined: 29 Oct 17
Posts: 1050
Credit: 16,614,827
RAC: 12,088
Message 68268 - Posted: 11 Feb 2023, 21:58:50 UTC - in response to Message 68267.  

The %age CPU time in boincmgr is set to ~50%. The difference between the wall-clock time is ~1min and the last CPU time given is 27secs, about a half:

  22:59:51 STEP  819 H= 819:00 +CPU= 42.573
  23:00:47 STEP  820 H= 820:00 +CPU= 27.309

I'd guess it was a laptop or mini-PC. I set mine at 60% otherwise it gets too warm if I'm running multiple tasks on it.
ID: 68268 · Report as offensive     Reply Quote
ProfileDave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 4541
Credit: 19,039,635
RAC: 18,944
Message 68272 - Posted: 12 Feb 2023, 7:22:59 UTC - in response to Message 68268.  

I'd guess it was a laptop or mini-PC. I set mine at 60% otherwise it gets too warm if I'm running multiple tasks on it.
I prefer to run fewer concurrent tasks if things start heating up rather than reduce the percentage of time cpu is running them.
ID: 68272 · Report as offensive     Reply Quote
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1120
Credit: 17,202,915
RAC: 2,154
Message 68284 - Posted: 12 Feb 2023, 22:35:23 UTC - in response to Message 68272.  

I prefer to run fewer concurrent tasks if things start heating up rather than reduce the percentage of time cpu is running them.


That is what I do too. Right now I would be running12 concurrent boinc tasks on my 8/16 core machine because it is cool here in my non-air-conditioned room. In summer, when it is hot, I reduce it to 8 or even 7 if it gets really hot.

Actually, I am running only 11 tasks because I have not gotten any Rosetta tasks in a very long time, and only one CPDN task every few days lately. So I am grinding out a lot of MilkyWay and Universe tasks that do not really interest me.
ID: 68284 · Report as offensive     Reply Quote
mikey

Send message
Joined: 18 Nov 18
Posts: 21
Credit: 6,635,794
RAC: 2,524
Message 68330 - Posted: 15 Feb 2023, 1:56:04 UTC - in response to Message 68284.  

I prefer to run fewer concurrent tasks if things start heating up rather than reduce the percentage of time cpu is running them.


That is what I do too. Right now I would be running12 concurrent boinc tasks on my 8/16 core machine because it is cool here in my non-air-conditioned room. In summer, when it is hot, I reduce it to 8 or even 7 if it gets really hot.

Actually, I am running only 11 tasks because I have not gotten any Rosetta tasks in a very long time, and only one CPDN task every few days lately. So I am grinding out a lot of MilkyWay and Universe tasks that do not really interest me.


There's always Asteroids or even MooWrapper or even Cosmology if you want to stick with 'space' projects. If you want to branch out a bit there are LOTS of other project that would love to see you help them too, even some that aren't in the Boinc Manager list of projects unless you've manually downloaded and updated it and even then some are still missing. ie Albert is the test project for Einstein, LHC has LHCdev, Rosetta has Ralph and PrimeGrid has Composite Grid as it's test project, There also basic projects like Gaia, Private GFN Server, SiDock, Yafu, TN-Grid and Wanless. Also Ithena has a 2nd project using comp instead of root at the beginning of it. Not ALL projects have both Windows and Linux tasks and the test projects are often out of work but you should be able to find something else that tickles your fancy in one or more of those.
ID: 68330 · Report as offensive     Reply Quote

Message boards : Number crunching : I feel sorry for this machine...

©2024 cpdn.org