Don't know where it should go? Stick it here!

Message boards : Number crunching : Don't know where it should go? Stick it here!
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 30 · 31 · 32 · 33 · 34 · 35 · 36 . . . 146 · Next

AuthorMessage
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1968994 - Posted: 6 Dec 2018, 18:12:51 UTC

Since the AMD Threadripper2 has 16 cores (32 threads) that are not directly connected to ram I wonder if it would be possible to compile a "small memory foot print" version of the cpu app that could fit all the data and the executable into the cpu cache?

I will freely admit, even if it is possible then we would have to come up with a way to run those only in those cpus unless they run virtually as fast as the regular version.

Tom
A proud member of the OFA (Old Farts Association).
ID: 1968994 · Report as offensive     Reply Quote
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1968995 - Posted: 6 Dec 2018, 18:15:41 UTC - in response to Message 1968737.  


BOINC will do exactly what you ask it to do, so if you ask it for a two day cache and your main is out, it will make a two day cache of your backup. As those deadlines will be different, it will likely go into EDF mode and crunch the backup until done! Unfortunately we can't set different caches for different projects.

Moral is, if you have a backup project(s) don't get greedy with the cache or it will fill with the backup!


Thank you for the reminder!

For my dedicated Seti machines I wish I could download enough tasks to get past a day long outage. Then I wouldn't be bothered by wanting to "keep them busy".

Tom
A proud member of the OFA (Old Farts Association).
ID: 1968995 · Report as offensive     Reply Quote
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1969340 - Posted: 8 Dec 2018, 16:15:34 UTC

I know that World Community Grid doesn't like to share with other Boinc projects. But it has lots of cancer and disease research projects. Is there another project with similar research that is "better behaved"?

Thank you.

Tom
A proud member of the OFA (Old Farts Association).
ID: 1969340 · Report as offensive     Reply Quote
Cruncher-American Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor

Send message
Joined: 25 Mar 02
Posts: 1513
Credit: 370,893,186
RAC: 340
United States
Message 1969342 - Posted: 8 Dec 2018, 16:22:31 UTC - in response to Message 1968737.  

BOINC will do exactly what you ask it to do, so if you ask it for a two day cache and your main is out, it will make a two day cache of your backup. As those deadlines will be different, it will likely go into EDF mode and crunch the backup until done! Unfortunately we can't set different caches for different projects.

Moral is, if you have a backup project(s) don't get greedy with the cache or it will fill with the backup!



I do not think that is quite correct, in the case when you set share 0 for your backup project. In that case, it will d/l only 1 WU at a time for the backup (at least, that's the way it used to be, back when I was doing backup projects).
ID: 1969342 · Report as offensive     Reply Quote
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1969370 - Posted: 8 Dec 2018, 18:53:46 UTC - in response to Message 1969342.  

BOINC will do exactly what you ask it to do, so if you ask it for a two day cache and your main is out, it will make a two day cache of your backup. As those deadlines will be different, it will likely go into EDF mode and crunch the backup until done! Unfortunately we can't set different caches for different projects.

Moral is, if you have a backup project(s) don't get greedy with the cache or it will fill with the backup!



I do not think that is quite correct, in the case when you set share 0 for your backup project. In that case, it will d/l only 1 WU at a time for the backup (at least, that's the way it used to be, back when I was doing backup projects).


Even when I set it to "0" Eienstein@home will regularly download as much as I asked for (0.5 days for instance).

However, World Community Gird sends a message that says "your settings don't allow us to send you anything" if I set it's priority to 0. :(

Tom
A proud member of the OFA (Old Farts Association).
ID: 1969370 · Report as offensive     Reply Quote
juan BFP Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 16 Mar 07
Posts: 9786
Credit: 572,710,851
RAC: 3,799
Panama
Message 1969372 - Posted: 8 Dec 2018, 19:01:05 UTC - in response to Message 1969370.  
Last modified: 8 Dec 2018, 19:02:31 UTC

Even when I set it to "0" Eienstein@home will regularly download as much as I asked for (0.5 days for instance).

Sorry, No you are wrong, If you set E@H to 0 (zero) it DL only 1 WU per GPU and only when the S@H cache is totally empty.
That's the way most of us work with a backup project.
World Community Grid does not allow you to seet it to 0 (zero) instead. So you can't use it as a backup project.
ID: 1969372 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1969381 - Posted: 8 Dec 2018, 20:16:45 UTC - in response to Message 1969372.  

Even when I set it to "0" Eienstein@home will regularly download as much as I asked for (0.5 days for instance).

Sorry, No you are wrong, If you set E@H to 0 (zero) it DL only 1 WU per GPU and only when the S@H cache is totally empty.
That's the way most of us work with a backup project.
World Community Grid does not allow you to seet it to 0 (zero) instead. So you can't use it as a backup project.

That doesn't match what I experience from Einstein. If set to "0", it will still download my global cache setting of 1 day. Einstein does not obey the BOINC rules on many things. I have to set NNT at all times and then Allow tasks when my cache gets low. It still sends too many for that single download and I have to abort 80-100 tasks immediately after resetting NNT. If I don't reduce my Einstein task cache below 200 or so, it will reduce my cache requests for Seti and not allow my standard Seti cache limit.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1969381 · Report as offensive     Reply Quote
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1969386 - Posted: 8 Dec 2018, 20:59:53 UTC - in response to Message 1969381.  

If I remember correctly that downloading more than 1 work unit per GPU was an error issue with BOINC. Can't remember where I saw it, Juan is correct that if BOINC is working correctly, then it only downloads 1 work unit per GPU. That said....I have had the issue that Keith mentions but haven't had it in a while. So, was it an issue with BOINC before 7.4.44??
ID: 1969386 · Report as offensive     Reply Quote
juan BFP Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 16 Mar 07
Posts: 9786
Credit: 572,710,851
RAC: 3,799
Panama
Message 1969389 - Posted: 8 Dec 2018, 21:40:19 UTC - in response to Message 1969381.  

That doesn't match what I experience from Einstein. If set to "0", it will still download my global cache setting of 1 day. Einstein does not obey the BOINC rules on many things. I have to set NNT at all times and then Allow tasks when my cache gets low. It still sends too many for that single download and I have to abort 80-100 tasks immediately after resetting NNT. If I don't reduce my Einstein task cache below 200 or so, it will reduce my cache requests for Seti and not allow my standard Seti cache limit.

You do something wrong, look my host and see, it only DL 1 WU per GPU when no Seti work is available.
Where you set the E@H preferences? Could be because the local preferences override the global ones.
ID: 1969389 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1969390 - Posted: 8 Dec 2018, 21:46:26 UTC - in response to Message 1969389.  

That doesn't match what I experience from Einstein. If set to "0", it will still download my global cache setting of 1 day. Einstein does not obey the BOINC rules on many things. I have to set NNT at all times and then Allow tasks when my cache gets low. It still sends too many for that single download and I have to abort 80-100 tasks immediately after resetting NNT. If I don't reduce my Einstein task cache below 200 or so, it will reduce my cache requests for Seti and not allow my standard Seti cache limit.

You do something wrong, look my host and see, it only DL 1 WU per GPU when no Seti work is available.
Where you set the E@H preferences? Could be because the local preferences override the global ones.

I only use global preferences and have always done so. That is up till just last night in my attempt to circumvent the BOINC bug that prevents any cpu task from running with my excluded gpus on the new TR host.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1969390 · Report as offensive     Reply Quote
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14650
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1969392 - Posted: 8 Dec 2018, 21:46:50 UTC - in response to Message 1969389.  

No, can't be local preferences - Resource Share isn't included in those settings. RS can only be set via the project web site - though you do have four venues to get confused between.
ID: 1969392 · Report as offensive     Reply Quote
Profile Siran d'Vel'nahr
Volunteer tester
Avatar

Send message
Joined: 23 May 99
Posts: 7379
Credit: 44,181,323
RAC: 238
United States
Message 1969398 - Posted: 8 Dec 2018, 22:50:52 UTC

Greetings,

I'm in a bit of a quandary about BOINCStats.

According to BOINCStats I have a total credits over all projects at 10Mil+. I looked at the values on my account page for all projects and it barely adds up to maybe 9Mil+. BOINCStats states I am second in my team.

My roommate has the #1 spot and I am just as confused about her totals too. The projects list on her account page barely adds up to 5Mil+ yet BOINCStats says she is at 24Mil+.

What gives? Am I reading all this wrong? What? I'm so confused....

Have a great day! :)

Siran
CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath
ID: 1969398 · Report as offensive     Reply Quote
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13722
Credit: 208,696,464
RAC: 304
Australia
Message 1969399 - Posted: 8 Dec 2018, 22:53:43 UTC - in response to Message 1969398.  
Last modified: 8 Dec 2018, 22:55:21 UTC

What gives? Am I reading all this wrong? What? I'm so confused....

Make sure you're looking at the right stats.
Seti main only, Seti main & Seti Beta, or all projects, with or without ASICs, etc.
Grant
Darwin NT
ID: 1969399 · Report as offensive     Reply Quote
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1969413 - Posted: 9 Dec 2018, 0:31:23 UTC - in response to Message 1969381.  

I have always seen Einstein send only 1 task when set to a RS of 0.
If it isn't 0 is when it gets aggressive.
ID: 1969413 · Report as offensive     Reply Quote
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1969448 - Posted: 9 Dec 2018, 4:58:34 UTC

I looked and several of us are running different version numbers of the Boinc Manager under Linux.

It maybe our experiences are different because out version numbers are different?

Tom
A proud member of the OFA (Old Farts Association).
ID: 1969448 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1969457 - Posted: 9 Dec 2018, 5:50:03 UTC - in response to Message 1969448.  

I looked and several of us are running different version numbers of the Boinc Manager under Linux.

It maybe our experiences are different because out version numbers are different?

Tom

Absolutely. The differences in the BOINC source code branches can be extreme. So no way to compare different versions of BOINC behaving exactly the same under the same circumstances.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1969457 · Report as offensive     Reply Quote
JohnDK Crowdfunding Project Donor*Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 28 May 00
Posts: 1222
Credit: 451,243,443
RAC: 1,127
Denmark
Message 1969518 - Posted: 9 Dec 2018, 16:09:24 UTC
Last modified: 9 Dec 2018, 16:11:15 UTC

I use BOINC 7.4.44 under Linux and 2x BOINC 7.14.2 under Win10 and a setting of 0 with Einstein, Milkyway and WCG works as it should, I only get work when there´s no SETI jobs and only 1 at a time.
ID: 1969518 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13161
Credit: 1,160,866,277
RAC: 1,873
United States
Message 1969571 - Posted: 9 Dec 2018, 18:17:45 UTC - in response to Message 1969518.  

I use BOINC 7.4.44 under Linux and 2x BOINC 7.14.2 under Win10 and a setting of 0 with Einstein, Milkyway and WCG works as it should, I only get work when there´s no SETI jobs and only 1 at a time.

Do you also run other projects concurrently on each host with Seti? I do and likely why I see the difference in behavior. I've learned long ago what works and what doesn't work with Einstein on my hosts. I might be a corner case. Seems likely since I am always finding ways to break BOINC as my recent problems with project_max_concurrent and exclude_gpus.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 1969571 · Report as offensive     Reply Quote
Profile Siran d'Vel'nahr
Volunteer tester
Avatar

Send message
Joined: 23 May 99
Posts: 7379
Credit: 44,181,323
RAC: 238
United States
Message 1969742 - Posted: 10 Dec 2018, 10:53:50 UTC - in response to Message 1969399.  
Last modified: 10 Dec 2018, 11:05:44 UTC

What gives? Am I reading all this wrong? What? I'm so confused....

Make sure you're looking at the right stats.
Seti main only, Seti main & Seti Beta, or all projects, with or without ASICs, etc.

Hi Grant,

All the numbers are for all projects worked or working to date. The numbers are not specific to one project or another. I'm trying to dig a little deeper to find a reason why the numbers are so out of whack.

Thanks Grant and have a great day! :)

Siran

[edit]
Ok, got it figured out. BOINCStats is adding in totals from retired projects and SETI account pages do not. So the numbers are correct.
[/edit]
CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath
ID: 1969742 · Report as offensive     Reply Quote
JohnDK Crowdfunding Project Donor*Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 28 May 00
Posts: 1222
Credit: 451,243,443
RAC: 1,127
Denmark
Message 1969744 - Posted: 10 Dec 2018, 11:13:22 UTC - in response to Message 1969571.  

I use BOINC 7.4.44 under Linux and 2x BOINC 7.14.2 under Win10 and a setting of 0 with Einstein, Milkyway and WCG works as it should, I only get work when there´s no SETI jobs and only 1 at a time.

Do you also run other projects concurrently on each host with Seti? I do and likely why I see the difference in behavior. I've learned long ago what works and what doesn't work with Einstein on my hosts. I might be a corner case. Seems likely since I am always finding ways to break BOINC as my recent problems with project_max_concurrent and exclude_gpus.

No I only run Seti.
ID: 1969744 · Report as offensive     Reply Quote
Previous · 1 . . . 30 · 31 · 32 · 33 · 34 · 35 · 36 . . . 146 · Next

Message boards : Number crunching : Don't know where it should go? Stick it here!


 
©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.