Multi core greater than 80 core


log in

Advanced search

Message boards : Number crunching : Multi core greater than 80 core

1 · 2 · 3 · 4 . . . 8 · Next
Author Message
Bry B
Send message
Joined: 3 Apr 99
Posts: 53
Credit: 832,165
RAC: 0
United States
Message 1085220 - Posted: 9 Mar 2011, 5:22:37 UTC

Having issues on my 160 LP system not being able to scale properly. I am only seeing aboutg 25% CPU utilization overall the processors. The processors that are being tagged are at 100% but its only using 25% of them. Are there optimized binaries available for large multi core systems? K group aware?

Profile HAL9000
Volunteer tester
Avatar
Send message
Joined: 11 Sep 99
Posts: 4664
Credit: 123,540,668
RAC: 98,669
United States
Message 1085227 - Posted: 9 Mar 2011, 5:58:02 UTC
Last modified: 9 Mar 2011, 6:03:38 UTC

I'm guessing that it is computer 5847405 as it shows only 40 CPU's. You might need to edit your account setting "On multiprocessors, use at most nn processors". I set the value to 256 so I wouldn't ever have a problem. As I have seem systems listed on the stats with 64 CPU's I don't think it should be to hard to get BOINC to see & use them all.

Edit: Todd Hebert has a 64 CPU, or at least 64 thread, server running. So if there is something besides the value I suggested I am sure he knows it.

Upon further though you might have to use the <ncpus>N</ncpus> in the cc_config.xml file.
____________
SETI@home classic workunits: 93,865 CPU time: 863,447 hours

Join the BP6/VP6 User Group today!

Bry B
Send message
Joined: 3 Apr 99
Posts: 53
Credit: 832,165
RAC: 0
United States
Message 1085230 - Posted: 9 Mar 2011, 6:15:38 UTC - in response to Message 1085227.
Last modified: 9 Mar 2011, 6:25:47 UTC

thanks,I checked and I am set already to 256 CPU's. I think I found part of the issue in which 64 GB memory had been removed from my lower tray. I had no idea. LOL I guess that makes sence since each processor in this archicture needs memory to move data. I will reconfig in the morning and see if that helps.

I can go up to 2tb in memory but I am not sure if this app can handle it. I think I will add some GPU's in the mix as well since it seems to like those. Anyone running large config hardware please advise. I will be going well beyond 160LP's very soon x64 windows.

PS: Any Windows IA-64 binaries out there? If so I could love to try this on my superdome 256 proc.

Thanks

Bry
____________

Profile HAL9000
Volunteer tester
Avatar
Send message
Joined: 11 Sep 99
Posts: 4664
Credit: 123,540,668
RAC: 98,669
United States
Message 1085231 - Posted: 9 Mar 2011, 6:21:25 UTC - in response to Message 1085230.

thanks,I checked and I am set already to 256 CPU's. I think I found part of the issue in which 64 GB memory had been removed from my lower tray. I had no idea. LOL I guess that makes sence since each processor in this archicture needs memory to move data. I will reconfig in the morning and see if that helps.

I can go up to 2tb in memory but I am not sure if this app can handle it. I think I will add some GPU's in the mix as well since it seems to like those. Anyone running large config hardware please advise. I will be going well beyond 320LP's very soon.

PS: Any Windows IA-64 binaries out there? If so I could love to try this on my superdome 256 proc.

Thanks

Bry


See the Lunatics "Alternate hardware platforms" section in their downloads area.

____________
SETI@home classic workunits: 93,865 CPU time: 863,447 hours

Join the BP6/VP6 User Group today!

Bry B
Send message
Joined: 3 Apr 99
Posts: 53
Credit: 832,165
RAC: 0
United States
Message 1085233 - Posted: 9 Mar 2011, 6:31:44 UTC - in response to Message 1085231.

awesome!! This is

I wonder what the largest single machine run was..... any idea?
____________

Profile BilBg
Volunteer tester
Avatar
Send message
Joined: 27 May 07
Posts: 2894
Credit: 6,616,356
RAC: 7,907
Bulgaria
Message 1085235 - Posted: 9 Mar 2011, 6:43:53 UTC - in response to Message 1085230.
Last modified: 9 Mar 2011, 6:48:04 UTC

I checked and I am set already to 256 CPU's.

This setting is old/obsolete (don't used by current BOINC)
Use/check (both on web and local (in BOINC Manager) preferences):
On multiprocessors, use at most 100% of the processors
http://setiathome.berkeley.edu/prefs.php?subset=global


I think I found part of the issue in which 64 GB memory had been removed from my lower tray. ...
I can go up to 2tb in memory but I am not sure if this app can handle it.


For SETI you need only ~30 MB per running task.

Also check:
Memory: when computer is in use, use at most XX% of total
Memory: when computer is not in use, use at most YY% of total

Edit:
For diagnosis it will be good to restart BOINC and post here the first 40-50 lines from Messages tab.
Some of us may spot the problem by looking at the messages.


____________



- ALF - "Find out what you don't do well ..... then don't do it!" :)

Bry B
Send message
Joined: 3 Apr 99
Posts: 53
Credit: 832,165
RAC: 0
United States
Message 1085247 - Posted: 9 Mar 2011, 8:56:21 UTC - in response to Message 1085235.

thanks,

here is some output:

3/9/2011 12:55:01 AM Starting BOINC client version 6.10.58 for windows_x86_64
3/9/2011 12:55:01 AM log flags: file_xfer, sched_ops, task
3/9/2011 12:55:01 AM Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3
3/9/2011 12:55:01 AM Data directory: C:\ProgramData\BOINC
3/9/2011 12:55:01 AM Running under account Administrator
3/9/2011 12:55:01 AM Processor: 60 GenuineIntel Intel(R) Xeon(R) CPU E7- 2870 @ 2.40GHz [Family 6 Model 47 Stepping 2]
3/9/2011 12:55:01 AM Processor: 256.00 KB cache
3/9/2011 12:55:01 AM Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 syscall nx lm vmx smx tm2 dca popcnt aes pbe
3/9/2011 12:55:01 AM OS: Microsoft Windows Server 2008 "R2": Enterprise x64 Edition, Service Pack 1, (06.01.7601.00)
3/9/2011 12:55:01 AM Memory: 63.99 GB physical, 127.98 GB virtual
3/9/2011 12:55:01 AM Disk: 136.60 GB total, 62.52 GB free
3/9/2011 12:55:01 AM Local time is UTC -8 hours
3/9/2011 12:55:01 AM No usable GPUs found
3/9/2011 12:55:01 AM SETI@home Found app_info.xml; using anonymous platform
3/9/2011 12:55:01 AM SETI@home URL http://setiathome.berkeley.edu/; Computer ID 5847405; resource share 100
3/9/2011 12:55:01 AM SETI@home General prefs: from SETI@home (last modified 08-Mar-2011 22:10:17)
3/9/2011 12:55:01 AM SETI@home Computer location: work
3/9/2011 12:55:01 AM SETI@home General prefs: no separate prefs for work; using your defaults
3/9/2011 12:55:01 AM Reading preferences override file
3/9/2011 12:55:01 AM Preferences:
3/9/2011 12:55:01 AM max memory usage when active: 32762.76MB
3/9/2011 12:55:01 AM max memory usage when idle: 58972.97MB
3/9/2011 12:55:01 AM max disk usage: 10.00GB
3/9/2011 12:55:01 AM don't use GPU while active
3/9/2011 12:55:01 AM (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
3/9/2011 12:55:01 AM Using proxy info from GUI
3/9/2011 12:55:01 AM Using HTTP proxy itgproxy:80
3/9/2011 12:55:01 AM SETI@home Started upload of 09se10ac.25092.3748.3.10.111_0_0
3/9/2011 12:55:01 AM SETI@home Started upload of 09se10ac.25092.7020.3.10.119_1_0
3/9/2011 12:55:01 AM GPUs have become unusable; disabling tasks
3/9/2011 12:55:01 AM SETI@home Restarting task ap_10dc10aa_B5_P1_00213_20110308_31408.wu_0 using astropulse_v505 version 505
3/9/2011 12:55:01 AM SETI@home Restarting task ap_10dc10aa_B4_P0_00318_20110308_01033.wu_0 using astropulse_v505 version 505
3/9/2011 12:55:01 AM SETI@home Restarting task ap_10dc10aa_B5_P1_00278_20110308_31408.wu_1 using astropulse_v505 version 505
3/9/2011 12:55:01 AM SETI@home Restarting task ap_10dc10aa_B5_P0_00314_20110308_16602.wu_0 using astropulse_v505 version 505
3/9/2011 12:55:01 AM SETI@home Restarting task ap_10dc10aa_B6_P1_00206_20110308_25599.wu_0 using astropulse_v505 version 505
3/9/2011 12:55:01 AM SETI@home Restarting task ap_10ja11aa_B1_P0_00203_20110308_28451.wu_1 using astropulse_v505 version 505
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.31730.15200.14.10.31_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 10dc10aa.8500.2112.11.10.192_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 10dc10aa.8500.2112.11.10.151_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 10dc10aa.8500.2112.11.10.138_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09se10ac.25092.14791.3.10.174_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09se10ac.25092.14791.3.10.156_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.31730.15609.14.10.206_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task ap_10dc10aa_B6_P0_00333_20110308_22250.wu_1 using astropulse_v505 version 505
3/9/2011 12:55:01 AM SETI@home Restarting task 09se10ac.25092.16018.3.10.59_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09se10ac.25092.16018.3.10.47_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.16836.15.10.91_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.16836.15.10.59_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.31730.16836.14.10.60_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 10dc10aa.8500.3748.11.10.183_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 10dc10aa.8500.3748.11.10.182_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.17245.15.10.164_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09ja11aa.11641.9065.16.10.239_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09ja11aa.11641.9065.16.10.197_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09ja11aa.11641.9065.16.10.173_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.17654.15.10.229_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.17654.15.10.208_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.17654.15.10.186_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.31730.17654.14.10.215_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.31730.17654.14.10.209_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09ja11aa.11641.9474.16.10.218_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 10dc10aa.8500.4975.11.10.103_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.18472.15.10.94_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.18472.15.10.92_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.5304.18472.15.10.107_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.31730.18472.14.10.116_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09oc10aa.31730.18472.14.10.114_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 10dc10aa.8500.5793.11.10.11_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09se10ac.25092.18472.3.10.5_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Restarting task 09se10ac.25092.18472.3.10.26_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 05se10aa.17814.49795.5.10.131.vlar_2
3/9/2011 12:55:01 AM SETI@home Starting task 05se10aa.17814.49795.5.10.131.vlar_2 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 05se10aa.17814.49795.5.10.128.vlar_2
3/9/2011 12:55:01 AM SETI@home Starting task 05se10aa.17814.49795.5.10.128.vlar_2 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 05se10aa.17814.49795.5.10.115.vlar_2
3/9/2011 12:55:01 AM SETI@home Starting task 05se10aa.17814.49795.5.10.115.vlar_2 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 05se10aa.17814.49795.5.10.112.vlar_2
3/9/2011 12:55:01 AM SETI@home Starting task 05se10aa.17814.49795.5.10.112.vlar_2 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 10dc10aa.8500.6202.11.10.229_0
3/9/2011 12:55:01 AM SETI@home Starting task 10dc10aa.8500.6202.11.10.229_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 10dc10aa.8500.6202.11.10.216_1
3/9/2011 12:55:01 AM SETI@home Starting task 10dc10aa.8500.6202.11.10.216_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09se10ac.25092.18881.3.10.163_0
3/9/2011 12:55:01 AM SETI@home Starting task 09se10ac.25092.18881.3.10.163_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09oc10aa.5304.19699.15.10.195_0
3/9/2011 12:55:01 AM SETI@home Starting task 09oc10aa.5304.19699.15.10.195_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09oc10aa.31730.19699.14.10.201_0
3/9/2011 12:55:01 AM SETI@home Starting task 09oc10aa.31730.19699.14.10.201_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09ja11aa.11641.11519.16.10.201_0
3/9/2011 12:55:01 AM SETI@home Starting task 09ja11aa.11641.11519.16.10.201_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting ap_10ja11aa_B2_P0_00149_20110308_11892.wu_1
3/9/2011 12:55:01 AM SETI@home Starting task ap_10ja11aa_B2_P0_00149_20110308_11892.wu_1 using astropulse_v505 version 505
3/9/2011 12:55:01 AM SETI@home Starting 10dc10aa.8500.7429.11.10.158_0
3/9/2011 12:55:01 AM SETI@home Starting task 10dc10aa.8500.7429.11.10.158_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09oc10aa.31730.20926.14.10.134_1
3/9/2011 12:55:01 AM SETI@home Starting task 09oc10aa.31730.20926.14.10.134_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09oc10aa.31730.20926.14.10.130_0
3/9/2011 12:55:01 AM SETI@home Starting task 09oc10aa.31730.20926.14.10.130_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09oc10aa.31730.20926.14.10.121_1
3/9/2011 12:55:01 AM SETI@home Starting task 09oc10aa.31730.20926.14.10.121_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09ja11aa.11641.12746.16.10.154_0
3/9/2011 12:55:01 AM SETI@home Starting task 09ja11aa.11641.12746.16.10.154_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 10dc10aa.8500.8247.11.10.63_0
3/9/2011 12:55:01 AM SETI@home Starting task 10dc10aa.8500.8247.11.10.63_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09oc10aa.5304.21744.15.10.92_1
3/9/2011 12:55:01 AM SETI@home Starting task 09oc10aa.5304.21744.15.10.92_1 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09oc10aa.31730.21744.14.10.22_0
3/9/2011 12:55:01 AM SETI@home Starting task 09oc10aa.31730.21744.14.10.22_0 using setiathome_enhanced version 603
3/9/2011 12:55:01 AM SETI@home Starting 09ja11aa.11641.13564.16.10.99_0
3/9/2011 12:55:01 AM SETI@home Starting task 09ja11aa.11641.13564.16.10.99_0 using setiathome_enhanced version 603

Profile soft^spirit
Avatar
Send message
Joined: 18 May 99
Posts: 6374
Credit: 28,647,395
RAC: 516
United States
Message 1085254 - Posted: 9 Mar 2011, 9:28:30 UTC
Last modified: 9 Mar 2011, 9:32:47 UTC

Okay.. it looks like you are spitting valid tasks out of that thing left and right.. Most are still waiting for validation. Should be interesting to see what that beast can do!!

Edit: I am guessing you have 4 10 core CPU's, which would give you 40 without
hyperthreading, and if available 80 with?

You might test both ways, as it might be no faster with hyperthreading enabled.
____________

Janice

Profile BilBg
Volunteer tester
Avatar
Send message
Joined: 27 May 07
Posts: 2894
Credit: 6,616,356
RAC: 7,907
Bulgaria
Message 1085256 - Posted: 9 Mar 2011, 9:43:00 UTC - in response to Message 1085247.


I see "Processor: 60 GenuineIntel Intel(R) Xeon(R) CPU E7- 2870 @ 2.40GHz"
and actually 60 tasks are running (Restarted + Started).

Is it what you expect, i.e.:
1) Do you think you have 60 (or more) virtual CPUs (according to the computer/CPU specs)?
2) Do you see (exactly) 60 virtual CPUs in Windows Task Manager -> Performance ?


Example picture of 128 virtual CPUs in Windows Task Manager:





____________



- ALF - "Find out what you don't do well ..... then don't do it!" :)

Bry B
Send message
Joined: 3 Apr 99
Posts: 53
Credit: 832,165
RAC: 0
United States
Message 1085264 - Posted: 9 Mar 2011, 10:41:33 UTC - in response to Message 1085254.

Okay.. it looks like you are spitting valid tasks out of that thing left and right.. Most are still waiting for validation. Should be interesting to see what that beast can do!!

Edit: I am guessing you have 4 10 core CPU's, which would give you 40 without
hyperthreading, and if available 80 with?

You might test both ways, as it might be no faster with hyperthreading enabled.



actually i have 8 10 core CPU's. Problem is that I noticed my contractor puled out memory on the lower tray. Ill get that fixed in the morning. so we have 80 without HT and with HT on its 160
____________

Bry B
Send message
Joined: 3 Apr 99
Posts: 53
Credit: 832,165
RAC: 0
United States
Message 1085265 - Posted: 9 Mar 2011, 10:48:31 UTC - in response to Message 1085256.

ok my first attempt to post a pic.


____________

Bry B
Send message
Joined: 3 Apr 99
Posts: 53
Credit: 832,165
RAC: 0
United States
Message 1085268 - Posted: 9 Mar 2011, 11:02:01 UTC - in response to Message 1085265.

I found some older binaries for IA-64 on windows. I loaded them on a dual socket to see how it would take. So far scaling looks good at 100% across the board. Ill get a new SP1 build of windows 2k8r2 on my 256 way tomorrow and see how that goes. Ill post task man pics. The X64 160 LP with only 80 working seems to be really chuggin along with over 60 tasks open processing.

This system above I thats 160 LP's I can comment on. Its one of our DL980 G7 systems. If your interested take a look at the specs at www.hp.com

Profile BilBg
Volunteer tester
Avatar
Send message
Joined: 27 May 07
Posts: 2894
Credit: 6,616,356
RAC: 7,907
Bulgaria
Message 1085269 - Posted: 9 Mar 2011, 11:03:25 UTC - in response to Message 1085265.


Try to set:
On multiprocessors, use at most 50% of the processors

to see what happens (will BOINC use 80 or 30?)


____________



- ALF - "Find out what you don't do well ..... then don't do it!" :)

Bry B
Send message
Joined: 3 Apr 99
Posts: 53
Credit: 832,165
RAC: 0
United States
Message 1085270 - Posted: 9 Mar 2011, 11:10:30 UTC - in response to Message 1085269.


Try to set:
On multiprocessors, use at most 50% of the processors

to see what happens (will BOINC use 80 or 30?)




i set this, suspended then update, then resume. Prior I was up to 37% with 40 of my CPU's pegged. The setting didnt seem to change anything. I am certian without memory in the lower tray attached to the second 4 sockets is the reason why the other processors wont register. Ill find out in a few hours when I get in.
____________

ClaggyProject donor
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4241
Credit: 34,944,761
RAC: 22,965
United Kingdom
Message 1085273 - Posted: 9 Mar 2011, 11:21:29 UTC - in response to Message 1085247.

thanks,

here is some output:

3/9/2011 12:55:01 AM Starting BOINC client version 6.10.58 for windows_x86_64
3/9/2011 12:55:01 AM log flags: file_xfer, sched_ops, task
3/9/2011 12:55:01 AM Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3
3/9/2011 12:55:01 AM Data directory: C:\ProgramData\BOINC
3/9/2011 12:55:01 AM Running under account Administrator
3/9/2011 12:55:01 AM Processor: 60 GenuineIntel Intel(R) Xeon(R) CPU E7- 2870 @ 2.40GHz [Family 6 Model 47 Stepping 2]
3/9/2011 12:55:01 AM Processor: 256.00 KB cache
3/9/2011 12:55:01 AM Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 syscall nx lm vmx smx tm2 dca popcnt aes pbe
3/9/2011 12:55:01 AM OS: Microsoft Windows Server 2008 "R2": Enterprise x64 Edition, Service Pack 1, (06.01.7601.00)
3/9/2011 12:55:01 AM Memory: 63.99 GB physical, 127.98 GB virtual
3/9/2011 12:55:01 AM Disk: 136.60 GB total, 62.52 GB free
3/9/2011 12:55:01 AM Local time is UTC -8 hours
3/9/2011 12:55:01 AM No usable GPUs found
3/9/2011 12:55:01 AM SETI@home Found app_info.xml; using anonymous platform
3/9/2011 12:55:01 AM SETI@home URL http://setiathome.berkeley.edu/; Computer ID 5847405; resource share 100
3/9/2011 12:55:01 AM SETI@home General prefs: from SETI@home (last modified 08-Mar-2011 22:10:17)
3/9/2011 12:55:01 AM SETI@home Computer location: work
3/9/2011 12:55:01 AM SETI@home General prefs: no separate prefs for work; using your defaults
3/9/2011 12:55:01 AM Reading preferences override file
3/9/2011 12:55:01 AM Preferences:
3/9/2011 12:55:01 AM max memory usage when active: 32762.76MB
3/9/2011 12:55:01 AM max memory usage when idle: 58972.97MB
3/9/2011 12:55:01 AM max disk usage: 10.00GB
3/9/2011 12:55:01 AM don't use GPU while active
3/9/2011 12:55:01 AM (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
3/9/2011 12:55:01 AM Using proxy info from GUI
3/9/2011 12:55:01 AM Using HTTP proxy itgproxy:80

3/9/2011 12:55:01 AM SETI@home Started upload of 09se10ac.25092.3748.3.10.111_0_0
3/9/2011 12:55:01 AM SETI@home Started upload of 09se10ac.25092.7020.3.10.119_1_0
3/9/2011 12:55:01 AM GPUs have become unusable; disabling tasks


According to your startup log, you've set this host to the 'work' venue,
but you don't have any 'work' preferences, so Boinc is using the default prefs,
but is then using the local preferences which over-ride Web preferences,
you could try Clearing the local preferences to see if more CPU's are utilised,

Claggy

Profile -= Vyper =-Project donor
Volunteer tester
Avatar
Send message
Joined: 5 Sep 99
Posts: 1097
Credit: 328,547,362
RAC: 144,702
Sweden
Message 1085285 - Posted: 9 Mar 2011, 12:14:29 UTC

Hmm

I have an idea, perhaps it's a limit in boinc to only handle 64 subprocesses and it can't load more cores by itself.

If you take a read at my blog i have a workaround that could suit the need to fully utilize that system.

Unfortunately the hosts get separate "hosts id" though they are running in the same machine but you can easily merge the numbers you get from the system to see the total RAC of that system.

Look here for the principals to run more than one boinc installation in one system.
http://vyper.kafit.se/wp/index.php/2011/02/04/running-different-nvidia-architectures-most-optimal-at-setihome/

Kind regards Vyper
____________

_________________________________________________________________________
Addicted to SETI crunching!
Founder of GPU Users Group

Profile -= Vyper =-Project donor
Volunteer tester
Avatar
Send message
Joined: 5 Sep 99
Posts: 1097
Credit: 328,547,362
RAC: 144,702
Sweden
Message 1085286 - Posted: 9 Mar 2011, 12:18:49 UTC

3/9/2011 12:55:01 AM Processor: 60 GenuineIntel Intel(R) Xeon(R) CPU E7- 2870 @ 2.40GHz [Family 6 Model 47 Stepping 2]


Boinc only seems to see 60 (cores physical as virtual) cores and thus cannot adress more cores than this! Whatever change you do if that numbers show others then you would most ceratinly succeed in utilizing more cores in the system when finding that magical tweak.

//Vyper
____________

_________________________________________________________________________
Addicted to SETI crunching!
Founder of GPU Users Group

Profile HelliProject donor
Volunteer tester
Avatar
Send message
Joined: 15 Dec 99
Posts: 705
Credit: 92,966,659
RAC: 60,727
Germany
Message 1085301 - Posted: 9 Mar 2011, 13:47:11 UTC

Whoa! A 10-Core CPU! With HyperThreading and 30MB L3 Cache. And then eight of them...
To bad that BOINC only supports 60 Cores. LOL

Maybe in a future Version of BOINC....


Helli
____________
A loooong time ago: My first Credits

Profile HAL9000
Volunteer tester
Avatar
Send message
Joined: 11 Sep 99
Posts: 4664
Credit: 123,540,668
RAC: 98,669
United States
Message 1085304 - Posted: 9 Mar 2011, 14:23:15 UTC - in response to Message 1085301.

Whoa! A 10-Core CPU! With HyperThreading and 30MB L3 Cache. And then eight of them...
To bad that BOINC only supports 60 Cores. LOL

Maybe in a future Version of BOINC....


Helli



I disagree. This machine is using 64. So 60 doesn't seem to be a limit at least for BOINC 6.10.58.
____________
SETI@home classic workunits: 93,865 CPU time: 863,447 hours

Join the BP6/VP6 User Group today!

Profile HelliProject donor
Volunteer tester
Avatar
Send message
Joined: 15 Dec 99
Posts: 705
Credit: 92,966,659
RAC: 60,727
Germany
Message 1085305 - Posted: 9 Mar 2011, 14:25:14 UTC - in response to Message 1085304.

Aah...yes. Todds Monster. I forgot. :-)

Then it must be a configuration Problem.

Helli
____________
A loooong time ago: My first Credits

1 · 2 · 3 · 4 . . . 8 · Next

Message boards : Number crunching : Multi core greater than 80 core

Copyright © 2014 University of California