Hi! I run a dualcore system. I finished my first two workunits, now BOINC downloaded one fresh one. Now my CPU usage is 50%.
Is there a way to get it up to 100% or get an additional work unit?
Project Settings, not overruled by local settings as follows:
On multiprocessors, use at most
Enforced by version 5.10 and earlier 2 processors
On multiprocessors, use at most
Enforced by version 6.1+ 99 % of the processors
Use at most
(Can be used to reduce CPU heat)
Enforced by version 5.6+ 99 percent of CPU time
Computer is connected to the Internet about every
(Leave blank or 0 if always connected.
BOINC will try to maintain at least this much work.) 10 days
Maintain enough work for an additional
Enforced by version 5.10+ 10 days

Thanks!

Please reply back with the first 20 or so lines in the Messages tab of the BOINC GUI (from when you start BOINC). This will give us more info that may help sort it out.

22.12.2008 16:11:20||Starting BOINC client version 6.4.5 for windows_intelx86
22.12.2008 16:11:20||log flags: task, file_xfer, sched_ops
22.12.2008 16:11:20||Libraries: libcurl/7.19.0 OpenSSL/0.9.8i zlib/1.2.3
22.12.2008 16:11:20||Data directory: F:\\Boinc
22.12.2008 16:11:20||Running under account Administrator
22.12.2008 16:11:20||Processor: 2 AuthenticAMD AMD Athlon(tm) 64 X2 Dual Core Processor 6400+ [x86 Family 15 Model 67 Stepping 3]
22.12.2008 16:11:20||Processor features: fpu tsc pae nx sse sse2 3dnow mmx
22.12.2008 16:11:20||OS: Microsoft Windows XP: Professional x86 Editon, Service Pack 2, (05.01.2600.00)
22.12.2008 16:11:20||Memory: 3.00 GB physical, 4.84 GB virtual
22.12.2008 16:11:20||Disk: 407.16 GB total, 249.60 GB free
22.12.2008 16:11:20||Local time is UTC +1 hours
22.12.2008 16:11:20||Not using a proxy
22.12.2008 16:11:20||CUDA devices found
22.12.2008 16:11:20||Coprocessor: GeForce 8600 GT (1)
22.12.2008 16:11:20|climateprediction.net|URL: http://climateprediction.net/; Computer ID: 853828; location: (none); project prefs: default
22.12.2008 16:11:20|Spinhenge@home|URL: http://spin.fh-bielefeld.de/; Computer ID: 150346; location: home; project prefs: default
22.12.2008 16:11:20||General prefs: from climateprediction.net (last modified 22-Dec-2008 15:02:34)
22.12.2008 16:11:20||Host location: none
22.12.2008 16:11:20||General prefs: using your defaults
22.12.2008 16:11:20||Preferences limit memory usage when active to 767.60MB
22.12.2008 16:11:20||Preferences limit memory usage when idle to 2916.90MB
22.12.2008 16:11:20||Preferences limit disk usage to 50.00GB
22.12.2008 16:11:20|climateprediction.net|Restarting task hadcm3ivolc_l0i7_2000_80_86001251_2 using hadcm3i version 604

Have you recently upgraded to 6.4.5? If so, what were you running before then?

It says use 99% of processors on a multiprocessor. Change that to 100% and see what happens.

22.12.2008 16:33:21||Starting BOINC client version 6.4.5 for windows_intelx86
22.12.2008 16:33:21||log flags: task, file_xfer, sched_ops
22.12.2008 16:33:21||Libraries: libcurl/7.19.0 OpenSSL/0.9.8i zlib/1.2.3
22.12.2008 16:33:21||Data directory: F:\\Boinc
22.12.2008 16:33:21||Running under account Administrator
22.12.2008 16:33:21||Processor: 2 AuthenticAMD AMD Athlon(tm) 64 X2 Dual Core Processor 6400+ [x86 Family 15 Model 67 Stepping 3]
22.12.2008 16:33:21||Processor features: fpu tsc pae nx sse sse2 3dnow mmx
22.12.2008 16:33:21||OS: Microsoft Windows XP: Professional x86 Editon, Service Pack 2, (05.01.2600.00)
22.12.2008 16:33:21||Memory: 3.00 GB physical, 4.84 GB virtual
22.12.2008 16:33:21||Disk: 407.16 GB total, 249.60 GB free
22.12.2008 16:33:21||Local time is UTC +1 hours
22.12.2008 16:33:21||Not using a proxy
22.12.2008 16:33:21||CUDA devices found
22.12.2008 16:33:21||Coprocessor: GeForce 8600 GT (1)
22.12.2008 16:33:21|climateprediction.net|URL: http://climateprediction.net/; Computer ID: 853828; location: (none); project prefs: default
22.12.2008 16:33:21|Spinhenge@home|URL: http://spin.fh-bielefeld.de/; Computer ID: 150346; location: home; project prefs: default
22.12.2008 16:33:21||General prefs: from climateprediction.net (last modified 22-Dec-2008 16:26:35)
22.12.2008 16:33:21||Host location: none
22.12.2008 16:33:21||General prefs: using your defaults
22.12.2008 16:33:21||Preferences limit memory usage when active to 767.60MB
22.12.2008 16:33:21||Preferences limit memory usage when idle to 2916.90MB
22.12.2008 16:33:21||Preferences limit disk usage to 50.00GB
22.12.2008 16:33:21|climateprediction.net|Restarting task hadcm3ivolc_l0i7_2000_80_86001251_2 using hadcm3i version 604

Changed to 100%. Usage of CPU is at 50% not changing.
Before I updated today there was the same effect.

I could perhaps understand why it wasn\'t downloading another model if long-term debt had built up for spinhenge. But I don\'t understand why CPU usage is only 50%. It should be 100% with one cpdn model and a spinhenge work unit.

I use spinhenge since today as I recognized the 50% load of CPDN. At starting with spinhenge I got only one WU and there I had too only 50% load. Since I have more WUs from spinhenge, boinc computes 2 spinhenge tasks same time and CPU load is 100%.
I tried to have one CPDN and one spinhenge task running at same time but I don´t know how...

Thanks again for troubleshooting

Perhaps someone with some better BOINC knowledge will come along to help out.

It looks like your problem is your memory usage preference. The idle side looks ok but the active side seems to be limiting memory usage excessively, only ~760mb available. If you increase that preference to 50% or higher you should be able to run 2 at a time.

---

BOINC WIKI

BOINCing since 2002/12/8

I already changed the memory settings with no changing effect. As I said if there are two tasks to perform the CPU load is 100%...
Quite confusing...

Could you please unhide your computers? This may let us see other details that are relevant.

In your BOINC Manager Projects tab, what does the Status column say for CPDN?

---

Cpdn news

You may look on my computer from now... I have the german BOINC so I don´t
know exactly what you mean on the project tab it says \"- - -\" on the tasks tab it says \"active\" for CPDN.
Spinhenge says \"communication delayed\" and on the other tab \"active, high priority\"
SETI says also \"comm delayed\" and on the other tab \"acitve 0.03 CPU, 1 CUDA\"

Status on the Tasks tab is where it says Active
I think that the reason you haven\'t received a second climate model, is that BOINC has decided that it has enough work at the moment.
Also, when you use the GPU for cuda processing, this also uses some cpu time.

It\'s possible that you just need to be patient for a few weeks, and let BOINC get on with it.

---

Backups: Here

Okay I thought that too, but it´s wasting 50% CPU time!
So I will not leave the PC running anymore when I have nothing to do.
I hope this will be patched by BOINC.
Thank you all for support.

Perhaps you\'ve inadvertently created a preferences override file on your computer, which has set the percentage of processors to 50%.
(Manager menu -> Advanced -> Preferences)

Also, as I said before, the cuda application uses a small percentage of one of the processors. (Only 1% or 2%, I think. Perhaps less.) Which may or may not show up under Task Manager.
There\'s a SETI FAQ here on their cuda apps.

---

Backups: Here

Hi again Götz

David Anderson who is the Big Boss of BOINC in Berkeley is saying on the boinc_dev email list:

The work-fetch policy in 6.4 and 6.5 is flawed; it can result in idle CPUs and/or GPUs. I don\'t think there\'s a quick fix for this problem; rather, the work fetch policy needs to be transformed.

The CPDN moderators have already said in the CPDN forum News threads (top of Number Crunching on this forum) that we do not at the moment recommend BOINC version 6.4, though our advice was given before David Anderson mentioned this specific BOINC defect.

Does anyone else think it might be a good idea for Götz to uninstall BOINC 6.4.5 and go back to v.6.2.19 which we know works very well for CPDN?

I see that Götz is only attached to CPDN and Spinhenge so I don\'t know whether the BOINC defect could affect his computer. On the other hand, his computer has a CUDA-capable graphics card that BOINC has detected, and I suspect that CUDA is part of the work fetch problem in BOINC 6.4.

---

Cpdn news

I concur, especially in consideration of David\'s post. No point in being on the \"bleeding edge\".

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.

I concur, especially in consideration of David\'s post. No point in being on the \"bleeding edge\".

I normally like to be on the \"bleeding edge\" but in this case if you are not actively trying to use a CUDA card to crunch SETI or GPUgrid I would recommend staying with the 6.2 version until the 6.6 version comes out.

---

BOINC WIKI

BOINCing since 2002/12/8

Thanks for your advices! I´ll think about downgrading.
Thanks again!

BOINC and CPDN both need alpha and beta testers on the bleeding edge so that new versions of BOINC and climate models with new features can be debugged and improved. The advice we give in the forum News threads is intended for ordinary members who want everything to work without complications.

Götz, if you do uninstall and downgrade, please let us know whether you then receive a new model. Downgrading you should allow you to keep your current model, though it\'s a good idea to back up the BOINC Data folder before changing version. Download 6.2.19 from here.

---

Cpdn news