Posts by red-ray

1) Message boards : Number crunching : Development BOINC 7.0.62 (Message 1235923)
Posted 24 May 2012 by Profile red-ray
Post:
The output from boinccmd --get_simple_gui_info seems to have changed since 7.00.25 and with 7.00.28 only includes active tasks. Is this intentional?

I used to use this to count how many WUs I had in my cache but as this no longer works I chaged to using --get_state.

I used to use --get_tasks, but switched to --get_simple_gui_info as BOINC 6.10 has --get_results rather than --get_tasks.

C:\Program Files\BOINC>boinccmd --version
boinccmd,  built from BOINC 7.0.28

C:\Program Files\BOINC>boinccmd --get_simple_gui_info | find "  state: " | find " " /c
2

C:\Program Files\BOINC>boinccmd --get_state | find "  state: " | find " " /c
578
2) Message boards : Number crunching : nVidia 301.42 WHQL Driver Released - DVI sleep fix (Message 1235067)
Posted 22 May 2012 by Profile red-ray
Post:
nVidia have finally released the WHQL driver which should have the DVI sleep fix.

http://www.nvidia.com/Download/index.aspx

So far it looks OK on http://setiathome.berkeley.edu/show_host_detail.php?hostid=6379711 which has GTX 680, GTX 460 and 8400 GS GPUs.
3) Message boards : Number crunching : 62 AP_V5 Left In The Field (Message 1232415)
Posted 15 May 2012 by Profile red-ray
Post:
Another one has been returned, but now there are 4 Results in progress rather than the expected 2!

'Returned' in that context could just mean yet another deadline passed. That - or any one of a number of other outcomes - will require another resend, and perhaps another 25 days before we can finally put this sequence to bed, if the resend has gone to somebody taking advantage of the newly unlimited caches.

Yes, but it came back in 186.88 hours so how can it be a deadline? I guess it could have been aborted. 3 I could understand, but what could cause 4? Two timeout resends?

Is there any way to get a list of who has the 4 AP WUs please?
4) Message boards : Number crunching : 62 AP_V5 Left In The Field (Message 1232407)
Posted 15 May 2012 by Profile red-ray
Post:
Another one has been returned, but now there are 4 Results in progress rather than the expected 2!
5) Message boards : Number crunching : Task takes too lomg error (Message 1232121)
Posted 14 May 2012 by Profile red-ray
Post:
This sort of thing is a pet hate of mine.

+1 (one who if very tempted to at least fix BOINC to allow for different GPU speeds).
6) Message boards : Number crunching : Task takes too lomg error (Message 1232113)
Posted 14 May 2012 by Profile red-ray
Post:
No, flops would not help because they can't be set differently for the different cards. If you halved the flops, it would only help for work already in cache. The servers would compensate and halve the rsc_fpops_est values for new work so you'd be right back where you started.

Thank you for this Joe. I keep getting told I need to set the FLOPS and when I do so it works for a while and then the ERR_RSC_LIMIT_EXCEEDED errors start again. I suspected this was happening, but was unsure.
7) Message boards : Number crunching : Task takes too lomg error (Message 1232106)
Posted 14 May 2012 by Profile red-ray
Post:
I ran mixed speed GPUs on http://setiathome.berkeley.edu/show_host_detail.php?hostid=6379672 without getting this issue for 3 months but over the past few weeks it is happening more and more. I have just written a small C program that updates all the <rsc_fpops_bound> values in the client_state.xml to be 25 rather than 10 times the <rsc_fpops_est> values which should resolve this, I just need to stop BOINC and run it every few days. It also resets the DCF to 1.0.

I think you can do similar with Fred’s rescheduler, but have never used it.

The real problem is that BOINC thinks all the GPUs in a system must be the same speed and unless this is fixed running mixed speed GPUs will have issues. The main one for me now is that it's impossible to get a stable DCF. I just hope set@home starts using the <dont_use_dcf/> facility.
8) Message boards : Number crunching : System issue ? (Message 1231624)
Posted 13 May 2012 by Profile red-ray
Post:
Do you allow your monitor(s) to go to sleep? The version of the driver you're using apparently has an issue with that. Could be what's causing your issue here?

I feel this is unlikely as with the driver DVI sleep issue you you get lots of errors like http://setiathome.berkeley.edu/show_host_detail.php?hostid=5851682 is getting.

It's fixed in the 301.24 driver, but that is a Beta release.
9) Message boards : Number crunching : Hello Computer 6137511, control your machine (Message 1231615)
Posted 13 May 2012 by Profile red-ray
Post:
http://setiathome.berkeley.edu/show_host_detail.php?hostid=5851682 looks like another Anonymous with an issue. Even the valid WUs struggled to get started.

I feel this system should be blocked.

I think we know that one:

NVIDIA GeForce GTS 450 (1024MB) driver: 296.10

Probably, but it should be blocked 'till the issue is resolved.
10) Message boards : Number crunching : System issue ? (Message 1231607)
Posted 13 May 2012 by Profile red-ray
Post:
This is probably a result of the limits being removed and wingmen taking longer. Is the pending count going up?

http://setiathome.berkeley.edu/results.php?hostid=5632153

My systems typically do 100K per day but today it's looking like it will be 85K and my pending has gone up by about 400.
11) Message boards : Number crunching : Hello Computer 6137511, control your machine (Message 1231591)
Posted 13 May 2012 by Profile red-ray
Post:
http://setiathome.berkeley.edu/show_host_detail.php?hostid=5851682 looks like another Anonymous with an issue. Even the valid WUs struggled to get started.

<stderr_txt>
Cuda error 'Couldn't get cuda device count
' in file 'c:/[Projects]/X_CudaMB/client/cuda/cudaAcceleration.cu' in line 146 : no CUDA-capable device is detected.
setiathome_CUDA: cudaGetDeviceCount() call failed.
setiathome_CUDA: No CUDA devices found
setiathome_CUDA: Found 0 CUDA device(s):
In cudaAcc_initializeDevice(): Boinc passed DevPref 1
setiathome_CUDA: CUDA Device 1 specified, checking...
   Device cannot be used
  Cuda device initialisation retry 1 of 6, waiting 5 secs...
Cuda error 'Couldn't get cuda device count

I feel this system should be blocked.
12) Message boards : Number crunching : Finally! GPU wars 2012 - GTX 650 Ti reviews (Message 1230721)
Posted 12 May 2012 by Profile red-ray
Post:
Got both of mine in today. Just going through my setup with Boinc to try and see what snags I may find. Do I need to turn SLi off when crunching?

No, my 2 x 460 are fine with SLI enabled. I also know of 3-way SLI 680s that work well.
13) Message boards : Number crunching : My new upgrade (Message 1230715)
Posted 11 May 2012 by Profile red-ray
Post:
Create the cc_config.xml file in the programdata\BOINC folder. It's needs to contain as below. Then exit and restart BOINC
<cc_config>
   <log_flags>
   </log_flags>
  <options>
    <use_all_gpus>1</use_all_gpus>
   </options>
</cc_config>
14) Message boards : Number crunching : My new upgrade (Message 1230712)
Posted 11 May 2012 by Profile red-ray
Post:
You need to include

<use_all_gpus>1</use_all_gpus>

in the options section of your cc_config.xml file.
15) Questions and Answers : Getting started : How do I defer results getting reported? (Message 1230523)
Posted 11 May 2012 by Profile red-ray
Post:
I have discovered this is a "feature" from changeset [trac]changeset:23571[/trac]. The word bug comes to mind.
16) Message boards : Number crunching : CUDA App Memory Usage (Message 1230502)
Posted 11 May 2012 by Profile red-ray
Post:
A lot of the "missing memory" is above 4GB and 32-bit Windows XP cannot use it. With 64-bit windows you could. On my system system with 12 GB and 4 GPUs the physical memory and GPU BARs are as below.



You should be able to use it with 32-Bit Windows Advanced/Enterprise Server, but I suspect you may get issues with the nVidia drivers. Were it my system I would install W7 x64.
17) Message boards : Number crunching : CUDA App Memory Usage (Message 1230466)
Posted 11 May 2012 by Profile red-ray
Post:
I suspect the memory is is the file system cache transition list. As I recall XP task manager considers this to be free. Have a look at what SIV reports for the file system cache.

18) Message boards : Number crunching : Panic Mode On (74) Server problems? (Message 1230458)
Posted 11 May 2012 by Profile red-ray
Post:
I still reckon something's not quite right.
I'm not getting as many "Project has no tasks available" messages as i was, but i'm still getting more than i usually do even when network traffic is maxed out. Given how (relatively) low the traffic has been i would expect to get hardly any, if any, such messages when requesting work.

Now there are no limits I expect many hosts are asking for and getting the entire of the feeder buffer. Getting WUs is going to be a problem 'till all the caches are full. I feel it would help a lot if the feeder could have a bigger buffer.

I am puzzled as to why the Result average turnaround is dropping though.
19) Message boards : Number crunching : Panic Mode On (74) Server problems? (Message 1230092)
Posted 10 May 2012 by Profile red-ray
Post:
If you stop BOINC and set a bigish duration_correction_factor you will just get CPU work for a while. The reason my 980X gets CPU WUs is the DCF jumps to 6 when a slow GPU finishes and the system just asks for CPU WUs 'till it drops.

Wow, the 980X hast just hit 4,000 WUs cached.
20) Message boards : Number crunching : Panic Mode On (74) Server problems? (Message 1230072)
Posted 10 May 2012 by Profile red-ray
Post:
I thought there was talk about that being corrected in the v7 client, but then there is the odd high/low work fetch system it uses.

No, I have 7.0.25 on my QX6700 and it's got the same problem, so having V7 does not help with this server issue.

I would like to see a bigger fifo so fewer requests are needed to replenish the cache.


Next 20


 
©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.