Same WU takes more time in Lunatics (ATI) than only CPU ?

Message boards : Number crunching : Same WU takes more time in Lunatics (ATI) than only CPU ?
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · Next

AuthorMessage
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1285950 - Posted: 20 Sep 2012, 18:44:09 UTC - in response to Message 1285737.  


The more proper way will be to set 'Use CPU no' here:
http://setiathome.berkeley.edu/prefs.php?subset=project

... but it will affect only SETI@home
You will not get more CPU tasks from SETI@home in the future.
CPU tasks you already have will remain and will be computed.


The way to stop CPU usage/tasks for entire BOINC:
<ncpus>0</ncpus>

http://boinc.berkeley.edu/wiki/Client_configuration
"
<ncpus>N</ncpus>
Act as if there were N CPUs; i.e. to simulate 2 CPUs on a machine that has only 1.
To use the number of available CPUs, set the value to -1
(was 0 which in newer clients really means zero to e.g. only allow GPU computing).
"

 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1285950 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1285970 - Posted: 20 Sep 2012, 19:44:28 UTC

Doesnt work, CPU continuing working.
By your command !!!
ID: 1285970 · Report as offensive
Profile Gundolf Jahn

Send message
Joined: 19 Sep 00
Posts: 3184
Credit: 446,358
RAC: 0
Germany
Message 1285991 - Posted: 20 Sep 2012, 20:45:35 UTC - in response to Message 1285970.  

Doesnt work, CPU continuing working.

What exactly doesn't work? There've been several suggestions.

Gruß,
Gundolf
ID: 1285991 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286073 - Posted: 21 Sep 2012, 0:08:39 UTC - in response to Message 1285991.  

Doesnt work, CPU continuing working.

What exactly doesn't work?

Wrote this in config file and nothing happens: "The way to stop CPU usage/tasks for entire BOINC:
<ncpus>0</ncpus>"
By your command !!!
ID: 1286073 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1286076 - Posted: 21 Sep 2012, 0:26:42 UTC - in response to Message 1286073.  
Last modified: 21 Sep 2012, 0:49:18 UTC


Post (copy here) the entire contents of cc_config.xml

Is the file name exactly cc_config.xml (Not cc_config.xml.txt | config.xml | config)?
Where you put the file? (copy the full folder name)

Did you read the link/page 'Client configuration' before making the edit?


E.g. my cc_config.xml looks like this (Do Not use any line from it if you don't understand what it does, this is just an example):
<cc_config>

   <log_flags>
      <benchmark_debug>1</benchmark_debug>
      <dcf_debug>1</dcf_debug>
      <checkpoint_debug>0</checkpoint_debug>
      <work_fetch_debug>0</work_fetch_debug>
   </log_flags>

   <options>
      <start_delay>13</start_delay>
      <ncpus>-1</ncpus>

      <max_file_xfers>20</max_file_xfers>
      <max_file_xfers_per_project>10</max_file_xfers_per_project>

      <save_stats_days>1100</save_stats_days>
      <max_stdout_file_size>5242880</max_stdout_file_size>

      <zero_debts>0</zero_debts>
   </options>

</cc_config>


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1286076 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286088 - Posted: 21 Sep 2012, 0:49:11 UTC

This is the cc_config file i have:

<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
<ncpus>0</ncpus>
<report_results_immediately>1</report_results_immediately>
<max_file_xfers>20</max_file_xfers>
</options>
</cc_config>
By your command !!!
ID: 1286088 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1286089 - Posted: 21 Sep 2012, 0:54:13 UTC - in response to Message 1286088.  
Last modified: 21 Sep 2012, 1:04:05 UTC


What about my other questions?
Is the file really used/recognized by BOINC?
Did you restart BOINC after edit?

On which Computer ID do you try this?:
http://setiathome.berkeley.edu/hosts_user.php?userid=8405154

What do you mean by "nothing happens"?


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1286089 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286094 - Posted: 21 Sep 2012, 1:12:31 UTC - in response to Message 1286089.  


What about my other questions?
Is the file really used/recognized by BOINC?
Did you restart BOINC after edit?

On which Computer ID do you try this?:
http://setiathome.berkeley.edu/hosts_user.php?userid=8405154

What do you mean by "nothing happens"?


In this one: AuthenticAMD
AMD C-60 APU with Radeon(tm) HD Graphics [Family 20 Model 2 Stepping 0]

21-09-2012 01:50:58 | | Starting BOINC client version 7.0.28 for windows_intelx86
21-09-2012 01:50:58 | | log flags: file_xfer, sched_ops, task
21-09-2012 01:50:58 | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
21-09-2012 01:50:58 | | Data directory: C:\ProgramData\BOINC
21-09-2012 01:50:58 | | Running under account Zapiao
21-09-2012 01:50:58 | | Processor: 2 AuthenticAMD AMD C-60 APU with Radeon(tm) HD Graphics [Family 20 Model 2 Stepping 0]
21-09-2012 01:50:58 | | Processor: 512.00 KB cache
21-09-2012 01:50:58 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 htt pni ssse3 cx16 syscall nx lm svm sse4a ibs skinit wdt page1gb rdtscp
21-09-2012 01:50:58 | | OS: Microsoft Windows 7: Starter x86 Edition, Service Pack 1, (06.01.7601.00)
21-09-2012 01:50:58 | | Memory: 749.54 MB physical, 2.69 GB virtual
21-09-2012 01:50:58 | | Disk: 100.00 GB total, 77.85 GB free
21-09-2012 01:50:58 | | Local time is UTC +1 hours
21-09-2012 01:50:58 | | ATI GPU 0: Loveland (CAL version 1.4.1417, 256MB, 223MB available, 88 GFLOPS peak)
21-09-2012 01:50:58 | | OpenCL: ATI GPU 0: Loveland (driver version CAL 1.4.1417 (VM), device version OpenCL 1.1 AMD-APP-SDK-v2.4 (650.9), 256MB, 223MB available)
21-09-2012 01:50:58 | SETI@home | Found app_info.xml; using anonymous platform
21-09-2012 01:50:58 | | Config: report completed tasks immediately
21-09-2012 01:50:58 | | Config: use all coprocessors
21-09-2012 01:50:58 | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 6765676; resource share 100
21-09-2012 01:50:58 | SETI@home | General prefs: from SETI@home (last modified 16-Sep-2012 19:06:44)
21-09-2012 01:50:58 | SETI@home | Computer location: home
21-09-2012 01:50:58 | SETI@home | General prefs: no separate prefs for home; using your defaults
21-09-2012 01:50:58 | | Reading preferences override file
21-09-2012 01:50:58 | | Preferences:
21-09-2012 01:50:58 | | max memory usage when active: 749.54MB
21-09-2012 01:50:58 | | max memory usage when idle: 749.54MB
21-09-2012 01:50:58 | | max disk usage: 1.00GB
21-09-2012 01:50:58 | | (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
21-09-2012 01:50:58 | | Not using a proxy

Nothing happens about the CPUs used, no indication of not using CPU

And always getting this error even after reseting the project
21-09-2012 01:54:50 | SETI@home | Task 29my12ag.30645.3360.12.10.56_1 exited with zero status but no 'finished' file
21-09-2012 01:54:50 | SETI@home | If this happens repeatedly you may need to reset the project.
21-09

By your command !!!
ID: 1286094 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1286478 - Posted: 22 Sep 2012, 0:30:35 UTC - in response to Message 1286094.  

21-09-2012 01:50:58 | | Config: report completed tasks immediately
21-09-2012 01:50:58 | | Config: use all coprocessors

OK, these lines show that (some) cc_config.xml is used.
(are you sure you did edit the one in C:\ProgramData\BOINC and not some copy)


Nothing happens about the CPUs used, no indication of not using CPU

For test - try to change the line to:
<ncpus>1</ncpus>

Don't forget to restart BOINC after edit.
(probably just re-read of the config file will be OK for this option but it will be better to restart BOINC)

Then you have to see this in the log:
Number of usable CPUs has changed from 2 to 1. Running benchmarks.
max CPUs used: 1


And always getting this error even after reseting the project
21-09-2012 01:54:50 | SETI@home | Task 29my12ag.30645.3360.12.10.56_1 exited with zero status but no 'finished' file
21-09-2012 01:54:50 | SETI@home | If this happens repeatedly you may need to reset the project.

This happens for everyone from time to time.
Usually caused by some 'heavy' program, high HDD load, lack of free RAM (paging) - very probable with 'Memory 750 MB + Windows 7', network problems
- all that makes the computer unresponsive (for 30 seconds).
Unless you get 100s of these messages per day no need to worry.

Do not reset the project - the advice is for users of stock apps (in case the stock .exe is broken the reset will download new stock .exe file, not the case for you: 'SETI@home | Found app_info.xml; using anonymous platform')


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1286478 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286484 - Posted: 22 Sep 2012, 0:48:13 UTC - in response to Message 1286478.  

It is working!!!!!!! I am only getting WUs for ATI. Thank you for your assistance BUT my AMD it inst working. Always getting that error. HEEEEEEEEEEEEEEEEEELP
By your command !!!
ID: 1286484 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1286489 - Posted: 22 Sep 2012, 1:33:39 UTC - in response to Message 1286484.  

It is working!!!!!!! I am only getting WUs for ATI. Thank you for your assistance BUT my AMD it inst working. Always getting that error. HEEEEEEEEEEEEEEEEEELP

How can anyone "HEEEEEEEEEEEEEEEEEELP" if you say just "Always getting that error"
And what you mean by "It is working for ATI" "BUT my AMD it inst working" (AMD and ATI is the same company now)


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1286489 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 1286493 - Posted: 22 Sep 2012, 1:41:49 UTC

it looks like hes killed a bunch of cpu WU's. I assume this was an install error. reinstall lunatics. make sure you choose SSE3 for AMD only


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 1286493 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286508 - Posted: 22 Sep 2012, 2:35:54 UTC - in response to Message 1286489.  

It is working!!!!!!! I am only getting WUs for ATI. Thank you for your assistance BUT my AMD it inst working. Always getting that error. HEEEEEEEEEEEEEEEEEELP

How can anyone "HEEEEEEEEEEEEEEEEEELP" if you say just "Always getting that error"
And what you mean by "It is working for ATI" "BUT my AMD it inst working" (AMD and ATI is the same company now)


Sorry I am talking about this pc: AuthenticAMD
AMD C-60 APU with Radeon(tm) HD Graphics [Family 20 Model 2 Stepping 0]
(2 processadores) that it is getting this error even after reseting the project:
21-09-2012 01:54:50 | SETI@home | Task 29my12ag.30645.3360.12.10.56_1 exited with zero status but no 'finished' file
21-09-2012 01:54:50 | SETI@home | If this happens repeatedly you may need to reset the project. I am tired of reseting and nothing. The WUs just stops after 1 minute.
By your command !!!
ID: 1286508 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286511 - Posted: 22 Sep 2012, 2:46:28 UTC - in response to Message 1286493.  

it looks like hes killed a bunch of cpu WU's. I assume this was an install error. reinstall lunatics. make sure you choose SSE3 for AMD only

I have the same config in this other pc: GenuineIntel
Intel(R) Core(TM) i3 CPU M 380 @ 2.53GHz [Family 6 Model 37 Stepping 5]
(4 processadores) and works fine. So why in the AMD i am getting that error?
By your command !!!
ID: 1286511 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22158
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1286553 - Posted: 22 Sep 2012, 7:46:02 UTC - in response to Message 1284605.  

First, slow down.

I see you've created a NEW PC with the same hardware. I don't see any instruction in this thread to do that. This new PC is ONLY set up for CPU crunching, not CPU plus GPU.
Second, each time BOINC sees a significant change in the hardware, or software it takes a wild guess at what the execution time for the tasks is going to be. Normally this guess is many times greater than that achieved (ten to twenty times isn't not uncommon - when I installed the GPU on htis computer it came up with over 200 hours for a task that took about 3 hours)). It takes BOINC a few completed and validated work units to get to a good estimate - a real pain I know, but be patient (at least S@H's guess are oveer estimates, some other projects come up with gross underestimates when they start up...)

Your other AMD based cruncher has dumped over 400WU. This one reports NO application configuration details. I guess this is one you've "removed from the gene pool".
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1286553 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286556 - Posted: 22 Sep 2012, 7:51:14 UTC - in response to Message 1286553.  
Last modified: 22 Sep 2012, 7:52:36 UTC

I am talking about this pc this AuthenticAMD
AMD C-60 APU with Radeon(tm) HD Graphics [Family 20 Model 2 Stepping 0]
(2 processadores) that it is getting this error even after reseting the project:
21-09-2012 01:54:50 | SETI@home | Task 29my12ag.30645.3360.12.10.56_1 exited with zero status but no 'finished' file
21-09-2012 01:54:50 | SETI@home | If this happens repeatedly you may need to reset the project. I just want to know why it is giving this error and how to fix it.
By your command !!!
ID: 1286556 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22158
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1286559 - Posted: 22 Sep 2012, 7:56:05 UTC
Last modified: 22 Sep 2012, 7:58:13 UTC

What is the ID number of the computer - the description fits TWO of your crunchers.






(Spelling typo corrected)
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1286559 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286560 - Posted: 22 Sep 2012, 7:59:26 UTC - in response to Message 1286559.  
Last modified: 22 Sep 2012, 8:00:40 UTC

It s ID: 6779714
By your command !!!
ID: 1286560 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22158
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1286562 - Posted: 22 Sep 2012, 8:10:38 UTC

Currently this cruncher is only showing you have the lunatics application installed for the CPU, and nothing for the GPU.
The guess at the run time may be far greater than the actual run time, ignore it for now. (I would expect an average MD work unit to take 2-4 hours, but the guess couple be 200hours - see my earlier post)


Let it run in this state for a short time, looking for errors. If it starts to produce lots of errors then SUSPEND the project rather than re-setting.
(Ignore the >400 that have timed out, they were SENT to you with a stupidly short deadline, this is a known server-side issue - you are "lucky" I had a dump of over 1500 of these one day while at work...)
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1286562 · Report as offensive
Zapiao
Volunteer tester

Send message
Joined: 29 Oct 01
Posts: 110
Credit: 122,278
RAC: 0
Portugal
Message 1286564 - Posted: 22 Sep 2012, 8:19:30 UTC - in response to Message 1286562.  

Currently this cruncher is only showing you have the lunatics application installed for the CPU, and nothing for the GPU.

How come??? I installed Multibeam for ATI dude.
By your command !!!
ID: 1286564 · Report as offensive
Previous · 1 · 2 · 3 · Next

Message boards : Number crunching : Same WU takes more time in Lunatics (ATI) than only CPU ?


 
©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.