Posts by Justin La Sotten

1) Message boards : Number crunching : Linux: v7 CUDA? (Message 1379957)
Posted 11 Jun 2013 by Profile Justin La Sotten
Post:
It's rather hard to weed through these posts trying to find information regarding Linux.

I see that there are no stock applications for v7 + CUDA for Linux i386/x86_64. Is this planned?
I used to get a version (Lunatics?) from the KWSN site, but obviously there isn't a new one just yet.

While trolling these forums, I also stumbled onto a post that said something about using the old x41g-cuda app from Lunatics because it works with the v7 WUs. Is this true too?

For the time being, I've disabled the Anonymous Platform stuff, and am using the stock apps (I wasn't getting v7 to work at all). And while this is good, I'd like to get back to processing WUs on my 2 GPUs.

Thanks in advance for any help/advice/critisims/hate!
2) Message boards : Number crunching : Not receiving regular CPU tasks (Message 1274577)
Posted 23 Aug 2012 by Profile Justin La Sotten
Post:

He does have the app for Enhanced work already.

By the way, the AP app is also Lunatics, made by Urs and mirrored on my site, along with making the packages.

The MB app is Crunch3rs.


Ahh thanks for the corrections.
3) Message boards : Number crunching : Not receiving regular CPU tasks (Message 1274571)
Posted 23 Aug 2012 by Profile Justin La Sotten
Post:
What are your cache settings? (Both of them)

Boinc will fill the fastest device first, if you have too large a cache setting, you'll never get CPU Wu's,

Claggy


Interesting, I have my cache settings super high.
Minimum work buffer: 4 days
Max additional work buffer: 5 days
Use at most: 50GB of disk space

Never had problems before, always had a full pool of regular CPU WUs and a nice hefty list of CUDA WUs.

I'll try setting this down to 1 day of work and see what happens. I'll trial and error my way through this.

Thanks!
4) Message boards : Number crunching : Not receiving regular CPU tasks (Message 1274491)
Posted 23 Aug 2012 by Profile Justin La Sotten
Post:
I have a machine that, up until very recently, was crunching away on 2 GPUs and 12 processors. However, for the past few day the machine is doing nothing but GPU WUs. When I watch the event log it says:

Requesting new tasks for CPU and NVIDIA
Scheduler request completed: got 0 new tasks
No tasks sent
No tasks are available for Astropulse v505
No tasks are available for SETI@home v7
No tasks are available for AstroPulse v6


However, it does download Nvidia WUs (I have a huge pool of WUs to work on that are CUDA only, it doesn't download these all the time).

Machine info:

Linux Mint 13
BOINC v7.0.33
  Arkayn   - AK_V8_linux64_ssse3
  Lunatics - setiathome_x41g_x86_64-pc-linux-gnu_cuda32
  Arkayn   - ap_6.01r546_avx_linux64



Anyone have any ideas?
5) Message boards : Number crunching : Linux 32-bit CUDA Client? (Message 1244556)
Posted 11 Jun 2012 by Profile Justin La Sotten
Post:
I've been scouring the boards trying to find this.
From what I gather, there once was a 32-bit Linux CUDA client but it had problems and was eventually pulled.

Just wondering if this is still the case and that I haven't looked over it.

I've looked at Lunatics site, as well as Arkayns site. I've found links to crunch3rs site, but they all seem to be dead.
6) Message boards : Number crunching : Generic crunching questions (Message 1243432)
Posted 8 Jun 2012 by Profile Justin La Sotten
Post:
A cuda card running under Linux will be used by several BOINC projects, but not by SETI unless you install a third-party app yourself. The project itself has no Linux/cuda applications shown on the application list.


Ahh thank you so much. This is it. I've gone ahead and installed the Lunatics Linux CUDA client. As soon as I start up BOINC, boom, there it is running the CUDA client.


All I need to figure out now is why it threw away my Astropulse and SETI@home Enhanced 6.03 apps.
I assume it has something to do with app_info.xml.
There wasn't a previous one, so I had no reference point to view a working app_info for those two applications.

I'll keep plugging away.

Thanks for your help.
7) Message boards : Number crunching : Generic crunching questions (Message 1243413)
Posted 8 Jun 2012 by Profile Justin La Sotten
Post:
Ok, I've gone ahead an set my preferences so that they aren't hidden anymore.
It doesn't appear to have updated just yet. Just a waiting game now.
8) Message boards : Number crunching : Generic crunching questions (Message 1243394)
Posted 8 Jun 2012 by Profile Justin La Sotten
Post:
I've just set up a new linux workstation.
The workstation has 2 video cards:

Nvidia Quadro 2000
Nvidia Quadro NVS 295

The Quadro 2000 is my primary card and is currently set up for my monitors.
I've just installed the other video card, but it isn't currently powering any monitors, just installed.

When I look at my BOINC event log I see that it knows about both video cards and it even knows that the 2nd card is "not used". However, I can't tell if BOINC is using either of these card for CUDA crunching.
How can I tell?

When I look at the list of tasks and I look at the "application" column, I see these unique apps:

AstroPulse v6 6.03
SETI@home Enhanced 5.28
SETI@home Enhanced 6.03

The event log has a few messages that look like this:

Requesting new tasks for CPU and NVIDIA

My event logs non-seti@home lines are here http://pastebin.com/FwVWrzqL for you viewing pleasure.

I have my preferences set to "use GPU always".
Could it be that there simply isn't any GPU tasks to do?
Sorry for these questions, this is my first workstation with any kind of good video card, I've always just run general CPU before (servers and the like).

I want to get the most out of this work horse, is there anything I could do? I've read a bit about optimized clients, will they give a boost? (I'll read the sticky thread about these in a few minutes)

Thanks in advance for any help!
9) Questions and Answers : Unix/Linux : Very slow (Message 383331)
Posted 31 Jul 2006 by Profile Justin La Sotten
Post:
I'm having the same problem.
Under windows xp i was averaging around 400-500 credits per day on a P4 3.0 HT proc with an average time of about an hour and a half.
Under Ubuntu 6.06 with SMP, i'm averaging around 200-300 with an average time of about 3 hours.
This is a pretty significant decrease.

I understand the angle rate effects processing times, however, i'm looking at the data over a period of time, the linux machine is in every case significantly slower in every single case, there hasn't been a single time where the processing is faster on the linux side than it is on the windows side.

Could there be any other reason for this?
My kernel is setup to understand SMP with HT.

Any help would be appreciated.
10) Questions and Answers : Web site : Delete Host: Lose Credit? (Message 322084)
Posted 1 Jun 2006 by Profile Justin La Sotten
Post:
Thank you very much for your quick reply. Must appreciated.
Is that Dr. Demento in your avatar?
11) Questions and Answers : Web site : Delete Host: Lose Credit? (Message 321967)
Posted 1 Jun 2006 by Profile Justin La Sotten
Post:
I have several hosts on my list that no longer are in service (haven't been for a long time). They have no pending results and have the option to delete.
If i delete a host, will i lose all the work that they have done? Each host has a few thousand credits, i don't want to lose them. Any ideas?





 
©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.