Posts by David Anderson (not *that* DA)


log in
1) Message boards : Number crunching : Gripes and Kudos IV (Message 1638417)
Posted 25 days ago by Profile David Anderson (not *that* DA)
Gripe: donated to Seti in January, no green star.
Unsure if thats really my fault or ?
Unsure who to contact.
No, it's not an earthshaking issue :-)
2) Message boards : Number crunching : SETI/BOINC Milestones [ v2.0 ] - XXVIII (Message 1637849)
Posted 27 days ago by Profile David Anderson (not *that* DA)
50 million boinc.
Nearly 24 million Seti.

Not getting much Seti GPU work. Oh well. :-)
3) Message boards : Number crunching : Invalid Host Messaging (Message 1627559)
Posted 14 Jan 2015 by Profile David Anderson (not *that* DA)
http://setiathome.berkeley.edu/workunit.php?wuid=1672054677
is another in the questionable series, this one is: 19ap11ad.22063.48525.140733193388041.12.99

Most of us processing that wu got invalid, but one task
is now marked inconclusive, so ... that's probably not
a good sign?
4) Message boards : Number crunching : Invalid Host Messaging (Message 1619490)
Posted 27 Dec 2014 by Profile David Anderson (not *that* DA)
While I don't know about all results fields, I note that
GPUs being bad often give spike count of 30 so
it appears there is a strong correlation between
the invalid-gpu-results.
Sometimes its various counts sum to 30.

Not random, in other words, but with
a fairly strong pattern.
5) Message boards : Number crunching : No GPU Units Under Linux (Message 1619456)
Posted 27 Dec 2014 by Profile David Anderson (not *that* DA)
Thanks for this thread. I failed to realize I needed opencl for AP!
Now installed opencl and boinc sees CUDA and opencl.
Ubuntu 14.04, driver 311.113
6) Message boards : Number crunching : Ubuntu 14.04 updates break cuda (Message 1617632)
Posted 23 Dec 2014 by Profile David Anderson (not *that* DA)
(was out of town a week, just now back).
Enabled Trusty-Backports so I could install
nvidia-modprobe. Installed it and ran it (executable
is the same name).
Made no visible difference in the output of
lsmod|grep nv
but after
sudo /etc/init.d/boinc-client restart
now boinc sees my GPUs and nvidia_uvm suddenly
shows up in the output of
lsmod|grep nv
!.

GPUs at work again!
See Ubuntu bug 1361207 for additional details if you care.
7) Message boards : Number crunching : Ubuntu 14.04 updates break cuda (Message 1614000)
Posted 14 Dec 2014 by Profile David Anderson (not *that* DA)
For now just living with it. No time
to deal with this.
Plus I could make things much worse... :-)

Anyway, thanks for the tips, ML1.
8) Message boards : Number crunching : Ubuntu 14.04 updates break cuda (Message 1613899)
Posted 14 Dec 2014 by Profile David Anderson (not *that* DA)
Switched to Noveau driver.
reboot
(Aside: the boot screen is much prettier with Noveau driver
than with nvidia driver, noveau giving blue screen with cute
xubuntu image and fancy progress indicator, nvidia drivers
give boring black/white screen with dots progress indicator)
Removed all nvidia (with synaptic).
reboot
installed nvidia-331
reboot

lsmod:

nvidia-uvm did not get loaded. Did modprobe which
loaded it, but its use count is zero...

No GPU seen by boinc. So this is not a case of
the upgrade mechanism getting mixed up by my specific
history on the the machine.

While Seti has had no CUDA work for me for a while
Einstein has a steady CUDA supply (now those are stuck).
Reports on Ubuntu suggest this bug is maybe fixed(?) for
next release but I see no sign it's been backported to 14.04. Yet.
9) Message boards : Number crunching : Ubuntu 14.04 updates break cuda (Message 1613440)
Posted 13 Dec 2014 by Profile David Anderson (not *that* DA)
Updated kernel etc today. Minor update
supposedly. Ubuntu 14.04. x86_64
nvidia gtx 760SC (2 of them)
Suddenly CUDA ceased working, boinc
cannot find GPUs. Ubuntu bug 1401350.
Non-working driver: 331.113.
10) Message boards : Number crunching : CLOSED*SETI/BOINC Milestones [ v2.0 ] - XXVII*CLOSED (Message 1567582)
Posted 5 Sep 2014 by Profile David Anderson (not *that* DA)
50 million for boinc, but Seti is not managing to
get as much of the computing as I would expect.
Solar panels providing the electricity!
11) Message boards : Number crunching : Who uses PCIe extenders; do they work? (Message 1523361)
Posted 1 Jun 2014 by Profile David Anderson (not *that* DA)
Emboldened by this thread I bought an extender from Hong Kong
($25 shipped) and a second GTX 760 SC and
made a holder (had to mod case of that machine
somewhat).

To my astonishment it is working.
Had the machine off most of the
day getting it assemble so it will take a
while to get to stable results.
Still only at about 50% of the PSU 860W rating.

Not getting any Seti GPU work at present, so the GPUs
are going to Einstein. But hopefully soon will get
more Seti GPU tasks.

So thanks to all for bringing this up and providing
advice. Would never have tried this otherwise.
12) Message boards : Number crunching : empty stderr or GPU 30 counts (Message 1522060)
Posted 28 May 2014 by Profile David Anderson (not *that* DA)
Here I mean in this forum, not this thread!
13) Message boards : Number crunching : empty stderr or GPU 30 counts (Message 1522059)
Posted 28 May 2014 by Profile David Anderson (not *that* DA)
Ok. I'll suggest they post here. Thanks.
14) Message boards : Number crunching : empty stderr or GPU 30 counts (Message 1521705)
Posted 28 May 2014 by Profile David Anderson (not *that* DA)
A few contributors have systems with GPUs that generate a
very high percent of Invalid: either
a) generate many empty-stderr tasks or
b) generate counts of 30 (or that add to 30)
Of course I noticed due to Inconclusive results we share.

It's bothering some of them --- they
responded im PMs saying so, but they don't know what to try and
I found it difficult to understand (from previous threads
that mentioned these things)
what to suggest for either of these issues.

So: suggestions for either sort of problem?
How they should proceed?
15) Message boards : Number crunching : CLOSED - SETI/BOINC Milestones [ v2.0 ] - XXVI - CLOSED (Message 1486837)
Posted 10 Mar 2014 by Profile David Anderson (not *that* DA)
I'm at 20,000,000 Seti and 30,000,000 combined
as of today...
16) Message boards : Number crunching : confused about boinc vs seti vs app_config.xml vs stock vs lunatics (Message 1473928)
Posted 7 Feb 2014 by Profile David Anderson (not *that* DA)
Advanced->Read Config Files

I get (dropping timestamps)
Suspending computation - user request
Resuming computation
Re-reading cc_config.xml
Not using a proxy
Config: GUI RPCs allowed from:
log flags: file_xfer, sched_ops,task


I don't see any sign of reading app_config.xml

app_config.xml contains stuff I gleaned
from the board earlier and may not be
correct or all I need to run MB and AP:
<app_config>
<app>
<name>astropulse_v6</name>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.4</cpu_usage>
</gpu_versions>
</app>
<app>
<name>setiathome_v7</name>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>.4</cpu_usage>
</gpu_versions>
</app>
</app_config>

Anything else I should have in there or anything
I should fix?

Ahhh. Bad access flags which were: rw-r----- (owned by root).
Changed to read-by-anyone and now on rereading config
the event log adds 'Found app_config.xml' to the output!

I don't have any Seti gpu work on hand, at the moment.
Thank you Fred! You got me thinking.
17) Message boards : Number crunching : confused about boinc vs seti vs app_config.xml vs stock vs lunatics (Message 1473904)
Posted 7 Feb 2014 by Profile David Anderson (not *that* DA)
I run recent Linux (Ubuntu 13.10) which identifies boinc as 7.2.7.

Stock seti (not Lunatics) which is using the nvidia GTX 760
fine. As is Einstein. Perhaps incorrectly (?) I did not delete the files
from the previous Lunatics 41z install (I did drain the
task list to empty before installing stock).
I renamed my Lunatics-based v6 app_info.xml so it is no longer
noticed.

But I was hoping to use app_config.xml in the seti project
under the boinc /var/lib/boinc-client/projects/seti* directory to
get a couple tasks running the gtx760 SC to see how that went.
But apparently app_config.xml is ignored with stock seti?

Sorry if this sounds confused, but I am confused.
I would like to get more
GPU tasks running (two would be nice, GPU is 3GB ram)
and I don't really understand the right way.

Delete old Lunatics and reinstall that?
(I would drain the task queues with 'no new tasks' button
first).

Or don't I need Lunatics for that control of the GPU with Seti?

Any hints would be much appreciated, and detailed instructions
even more appreciated.
18) Message boards : Number crunching : BOINC produces about 70% of INVALID results (Message 1471403)
Posted 31 Jan 2014 by Profile David Anderson (not *that* DA)
You mention 8 virtual cores so I guess that is 4 cores with
hyperthreading on (set by BIOS) so Linux counts 8.
I never could figure out if
hyperthreading helps with boinc.
Or if it hurts or is irrelevant.
In the end I sort of worried that Linux would not
clearly distinguish whether tasks
were sharing a real core when that was
pointless (if that even makes any sense)
so I turned off Hyperthreading in my BIOS.
Thinking (probably incorrectly) that it could be
better to run on distinct real cores rather than
sharing a real core across 2 tasks.

I agree with Rob that it's
good to set the cpu time % at 100. It means 100% of the
time of the cores-you-allow, but I guess that is obvious.
With those 4 real cores I would think running 100% on one
would be unobtrusive.

I'm not that knowledgeable, really, so I should shut up
and hope you get useful advice from...someone.
19) Message boards : Number crunching : BOINC produces about 70% of INVALID results (Message 1465608)
Posted 18 Jan 2014 by Profile David Anderson (not *that* DA)
Boinc Manager (on Linux) has (in Tools-> Computing Preferences) both

On multiprocessor systems use at most ___ % of the processors

and

Use at most ___% cpu time.

It's not clear from the posts which control you have been using.
Set the multiprocessor % to 50 as a start.
Leave the Use-at-most field at 100

That way you will get half the cores (whether you have 2 or 4)
running boinc stuff. I find that the default GPU setup uses
a small percent of an additional core.

Not that this is necessarily going to help...
Oh. On one machine I was having flaky issues (GPU invalids, 30 spikes)
and replacing the power supply, replacing the GPU, and fixing the heat
sink on the Northbridge chip got things working ok again. Dunno whether it
was one or all of those that mattered. Saw some notes on the web
about PSUs that
are ok except have short glitches ...
which got me to replace the PSU.
20) Message boards : Number crunching : It tells me that i do not have a usible version of seti at home. (Message 1463221)
Posted 11 Jan 2014 by Profile David Anderson (not *that* DA)
Donald is now up and running boinc. He just needed
a little Linux help. Recent Ubuntu defaults to
a window manager that looks tablet-y, which many
find annoying and difficult (I dislike
that too, on my desktop anyway). Switching to xfce
as window manager helped.


Next 20

Copyright © 2015 University of California