Best app for a Xeon X5650 cpu in Linux

Message boards : Number crunching : Best app for a Xeon X5650 cpu in Linux
Message board moderation

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1914491 - Posted: 21 Jan 2018, 23:18:37 UTC - in response to Message 1914487.  

Probably a question for one of the old-timers like Richard Haselgrove or Rob Smith. They understand and know all the underpinnings of BOINC.

No one knows all about Boinc.


Richard gets pretty close.

Anyway, that different stderr is because not only BOINC was installed from repository but also Seti MB app. It's got some changes that adds those extra lines.

Interesting. So how far back in revisions does that MB app go. No version level listed in the stderr.txt output. It does have a strange name and I noticed that. Didn't make the connection that it isn't a current MB app. It has to be one that is formatted to understand MB8 though of course. How did it end up in the repository? I only thought that BOINC versions are in the repository.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1914491 · Report as offensive
Profile Jim-R.
Volunteer tester
Avatar

Send message
Joined: 7 Feb 06
Posts: 1494
Credit: 194,148
RAC: 0
United States
Message 1914571 - Posted: 22 Jan 2018, 13:21:33 UTC - in response to Message 1914491.  

I happened to recall installing that app from the repository so I went and looked it up. It's name in the repository is 'boinc-app-seti-graphics' and version 8.00~svn3701-1.

And thanks for the tip on the process app. I've used 'task manager' to try to track it down but the only thing I've been able to find out with that is that each wu I'm crunching takes a total of 4% of the cpu. I could not get it to glitch while monitoring with task manager in the foreground, but as soon as I pulled Firefox back up to type here it glitched again! This might seem to lead to Firefox itself, but I run the exact same version of Firefox with the exact same addons on every computer I have and this does not happen on the others. And it seems to happen any time I have a program open and actively waiting for keyboard input. (Just used Thunderbird to reply to an email and it froze.)

I'm going to let things run as they are until tonight then shut it down and change over to the optimized app. then we can see if the stderr.txt returns to normal (it should) and I'll check while it's down to see if my freezes are due to BOINC itself. (As I mentioned, I have paused all SETI crunching and it didn't go away so it's not SETI itself.)
Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.
ID: 1914571 · Report as offensive
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1914594 - Posted: 22 Jan 2018, 16:45:11 UTC

Just Googled your app version. This is what it has to say.
Note that this package has been primarily created for users of architectures for which SETI@home does not provide its application. If your architecture is x86 or AMD64 the BOINC client automatically downloads the latest SETI@home application if you participate in this project. There is no need to install this package then, except for it to take all technical hurdles from you to have your very custom SETI patch or new compiler flags evaluated. The configuration of BOINC to find the local SETI binary is all performed by the package.


So it looks like it pulled the BOINC nightly tarball from the time the package was put into the repository. Also see it marked as unstable. Notice the last line where it says it will do the configuration with your custom SETI patch and new compiler flags. The custom compiler flags are probably what prints out the strange stderr.txt output.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1914594 · Report as offensive
Profile Jim-R.
Volunteer tester
Avatar

Send message
Joined: 7 Feb 06
Posts: 1494
Credit: 194,148
RAC: 0
United States
Message 1914604 - Posted: 22 Jan 2018, 17:32:11 UTC - in response to Message 1914594.  

Ok, that seems to clear that up. I just came in for dinner and have to go back out on another call in a few minutes but that app will be gone tonight when I change over to the SSE4.2 optimized app, so hopefully, after tonight there will be no more weird entries! hehe.
Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.
ID: 1914604 · Report as offensive
Juha
Volunteer tester

Send message
Joined: 7 Mar 04
Posts: 388
Credit: 1,857,738
RAC: 0
Finland
Message 1914620 - Posted: 22 Jan 2018, 19:40:01 UTC - in response to Message 1914594.  

The custom compiler flags are probably what prints out the strange stderr.txt output.


Well, no. They take the revision this or that and apply some patches on top of it. One of the patches adds the messages. The compiler flags are for when there is a brand new compiler and you want to know if enabling SSE7 would make the app go any faster.
ID: 1914620 · Report as offensive
Previous · 1 · 2

Message boards : Number crunching : Best app for a Xeon X5650 cpu in Linux


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.