All things Nvidia GPU

Stephen "Heretic"
Stephen "Heretic"
Joined: 5 Feb 17
Posts: 94
Credit: 645067679
RAC: 0

San-Fernando-Valley

San-Fernando-Valley wrote:

... didn't want to "create" work for you ...

Maybe I'm getting to old for this kind of fiddeling around.

I tend to believe what you are saying, sometimes I have trouble remembering the basics ...

What do I conclude out of this?

Well, NO MORE POSTS from me, I guess !

Cheers

 . . Just my 2c worth, not having installed a new BOINC for quite a while, is it possible that during the setup there was a selection between local preferences and network preferences and you chose local?

Stephen

 

Stephen "Heretic"
Stephen "Heretic"
Joined: 5 Feb 17
Posts: 94
Credit: 645067679
RAC: 0

petri33 wrote:Hi,my

petri33 wrote:

Hi,

my normal runtime for a task is 124-179 seconds depending on which GPU it is run on.

Tonight I got a task that took 1200+ seconds to finish.

Are they "candidates" to be verified or processing errors? Why the extra time?

--

petri33

 

 

Task 1191519630

 

Name: LATeah4013L01_1124.0_0_0.0_9033365_1

Workunit ID: 587140640

Created: 14 Nov 2021 21:43:52 UTC

Sent: 14 Nov 2021 21:58:15 UTC

Report deadline: 28 Nov 2021 21:58:15 UTC

Received: 15 Nov 2021 10:33:39 UTC

Server state: Over

Outcome: Success

Client state: Done

Exit status: 0 (0x00000000)

Computer: 12836077

Run time (sec): 1,266.08

CPU time (sec): 269.10

Peak working set size (MB): 420.33

Peak swap size (MB): 19050.06

Peak disk usage (MB): 0.02

Validation state: Valid

Granted credit: 3,465

Application: Gamma-ray pulsar binary search #1 on GPUs v1.28 (FGRPopencl2Pup-nvidia)
x86_64-pc-linux-gnu


Stderr output

<core_client_version>7.16.6</core_client_version>
<![CDATA[
<stderr_txt>
12:12:26 (26533): [normal]: This Einstein@home App was built at: Aug 17 2021 16:19:40

12:12:26 (26533): [normal]: Start of BOINC application '../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.28_x86_64-pc-linux-gnu__FGRPopencl2Pup-nvidia'.
12:12:26 (26533): [debug]: 1e+16 fp, 6.1e+09 fp/s, 1710426 s, 475h07m05s57
12:12:26 (26533): [normal]: % CPU usage: 1.000000, GPU usage: 0.130000
command line: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.28_x86_64-pc-linux-gnu__FGRPopencl2Pup-nvidia --inputfile ../../projects/einstein.phys.uwm.edu/LATeah4013L01.dat --alpha 0.943218186562 --delta 1.30995332125 --skyRadius 2.617990e-08 --ldiBins 30 --f0start 1116.0 --f0Band 8.0 --firstSkyPoint 0 --numSkyPoints 1 --f1dot -1e-13 --f1dotBand 1e-13 --df1dot 1.713401242e-15 --ephemdir
</stderr_txt>
]]>




"

 . . Hi Petri,

 . . Do you normally run 8sies?  (My term for 8 concurrent tasks).

 . . 8 x 124 = 992; 8 x 179= 1432.  OR 1224 / 8 = 153 ... ???

Stephen

 

Keith Myers
Keith Myers
Joined: 11 Feb 11
Posts: 4893
Credit: 18430749077
RAC: 5737994

Petri stated his normal

Petri stated his normal runtime is around 120 seconds so he is running only 1X per gpu or maybe possibly 2X if he is running a new bit of tweaked code.

The fact that one ran 1200 seconds is anomaly that indicates the task got stuck somehow where the cpu couldn't service it correctly or that BOINC for some reason kept taking focus away from the task instead of processing it normally.

 

Stephen "Heretic"
Stephen "Heretic"
Joined: 5 Feb 17
Posts: 94
Credit: 645067679
RAC: 0

Keith Myers wrote: Petri

Keith Myers wrote:

Petri stated his normal runtime is around 120 seconds so he is running only 1X per gpu or maybe possibly 2X if he is running a new bit of tweaked code.

The fact that one ran 1200 seconds is anomaly that indicates the task got stuck somehow where the cpu couldn't service it correctly or that BOINC for some reason kept taking focus away from the task instead of processing it normally.

 . . Yep. I felt sure he normally runs onesies.  But that task ran as an 8sie (see the line I highlighted in the message).  I won't even try to guess why but it did ...

Stephen

 

Keith Myers
Keith Myers
Joined: 11 Feb 11
Posts: 4893
Credit: 18430749077
RAC: 5737994

Good catch, I missed that in

Good catch, I missed that in your quote. If you look at his current set of valid tasks, he is running onesies as you call it.

How that one task got switched to eightsies 0.13 gpu resource is a mystery.

All I can think of is he was editing his app_info and had a case of fat fingers and entered the anomalous value.

Then corrected it.  But if that was the case he would have known of the error and not posted the message about "what happened here"

 

Stephen "Heretic"
Stephen "Heretic"
Joined: 5 Feb 17
Posts: 94
Credit: 645067679
RAC: 0

Keith Myers wrote:Good

Keith Myers wrote:

Good catch, I missed that in your quote. If you look at his current set of valid tasks, he is running onesies as you call it.

How that one task got switched to eightsies 0.13 gpu resource is a mystery.

 . . Is that the theme from Twilight Zone I hear in the background. ... ???

:)

Stephen

 

PS: Have you noticed a change in run times on tasks sent out since about 2am today UTC?? On both of my serious crunchers run times have dropped from around 10 mins to about 8.5 mins consistently. 

<scratching head>

 

Stephen "Heretic"
Stephen "Heretic"
Joined: 5 Feb 17
Posts: 94
Credit: 645067679
RAC: 0

 . . I just had a peek at

 . . I just had a peek at your machine with the 2070s and the same thing is happening on it. Run times were 298 and 278 for the two cards, but now they are 262 and 246.  Something has changed. I wondered if maybe there had been an unrequested upgrade of the video drivers but not when I checked my machines.

Stephen

<head scratching furiously now>

 

 

petri33
petri33
Joined: 4 Mar 20
Posts: 123
Credit: 3704515819
RAC: 7128204

Thank you Stephen and Keith!

Thank you Stephen and Keith! I did a test and ran 8 tasks per GPU at a time. It did not yield to any speed-up. I had forgotten that and did not check the GPU % (0.13).

I  noticed that the task times have dropped: 15 sec off on my TITAN V and about 17 sec off on my RTX 2080 Ti

--

Petri

Ian&Steve C.
Ian&Steve C.
Joined: 19 Jan 20
Posts: 3910
Credit: 43665195976
RAC: 63143821

petri33 wrote: I  noticed

petri33 wrote:

I  noticed that the task times have dropped: 15 sec off on my TITAN V and about 17 sec off on my RTX 2080 Ti

--

Petri


 

yeah looks like we’re running the faster 3011 tasks. The 4011 series tasks we’ve been on for a while are a bit slower.

_________________________________________________________________________

Keith Myers
Keith Myers
Joined: 11 Feb 11
Posts: 4893
Credit: 18430749077
RAC: 5737994

Hmm,  no sign of fast running

Hmm,  no sign of fast running 3011's here.  Just the usual 4013's

[Edit] Damn Einstein's sorting preference.  See them now.

 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.