Panic Mode On (76) Server Problems?


log in

Advanced search

Message boards : Number crunching : Panic Mode On (76) Server Problems?

Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · 15 . . . 21 · Next
Author Message
Profile SliverProject donor
Avatar
Send message
Joined: 18 May 11
Posts: 281
Credit: 7,162,769
RAC: 7,496
United States
Message 1273778 - Posted: 21 Aug 2012, 22:48:32 UTC - in response to Message 1273777.
Last modified: 21 Aug 2012, 22:48:48 UTC

Possess it if you can.
Found seldom in a woman,
And never in a man!
(From my mother's autograph book, ca. 1935)


LOL! That was good, ivan :D
____________

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,497,555
RAC: 25,398
United States
Message 1273779 - Posted: 21 Aug 2012, 22:48:36 UTC - in response to Message 1273776.

I dont talk about this.

But you did.

VLARs will not be sent to nvidia hosts.
End of story.


Until the next time they get sent to Nvidia cards.
It's happened before. It will probably happen again.

And a good day to you too.

Profile Sunny129
Avatar
Send message
Joined: 7 Nov 00
Posts: 190
Credit: 3,163,755
RAC: 0
United States
Message 1273782 - Posted: 21 Aug 2012, 22:53:46 UTC - in response to Message 1273772.
Last modified: 21 Aug 2012, 22:55:37 UTC

Then you should define what "most" is/means
because my Nvidias from the GTX/GTS 460/450
on up have always been able to run them.

Not as fast as some might like, but work is work.

Unless one is into cherry picking. Don't know what
else it can be called. Flushing some work units that can
be crunched because they run slower than other work units
is cherry picking in my book.

If they cause errors or cause too great a lag
so that it interferes with what the PC is normally
used for, then one has a valid reason to abort them.

Just because they run "slow" is not one of them,IMHO.

to elaborate on what Mike is saying about "most" nVidia GPU hosts not receiving VLARs, i believe that the few nVidia GPU hosts out there that do occasionally (or regularly) receive VLARs are receiving them in error. you see, the server is coded to take angle range into account and prevent VLARs from going out to nVidia GPU hosts. i don't know if the error lies within the server or the handful of nVidia GPU hosts that accidentally receive VLARs.

*EDIT* - Bill, out of curiosity, when was the last time you saw a VLAR running on one of your nVidia GPUs?
____________

Profile Wiggo
Avatar
Send message
Joined: 24 Jan 00
Posts: 6949
Credit: 94,621,878
RAC: 75,224
Australia
Message 1273783 - Posted: 21 Aug 2012, 22:54:10 UTC - in response to Message 1273779.

My GTX560 Ti & dual GTX550 Ti rigs don't mind the VLARs at all but overnight my triple 9800 rig started on its lot with very bad results eventually locking it up while I slept, so they all got aborted this morning on that setup.

Cheers.
____________

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,497,555
RAC: 25,398
United States
Message 1273784 - Posted: 21 Aug 2012, 23:01:43 UTC - in response to Message 1273782.

<snippage occured>

*EDIT* - Bill, out of curiosity, when was the last time you saw a VLAR running on one of your nVidia GPUs?[/quote]

3 are running concurrently right now. Ran out of regular
gpu work units during the outage so shipped a few over from the cpu.

Why? And yes I did receive some for the Nvidia gpu during the last server hiccup. Did those too on the gpu. Run times were posted in an earlier post.

Profile Sunny129
Avatar
Send message
Joined: 7 Nov 00
Posts: 190
Credit: 3,163,755
RAC: 0
United States
Message 1273789 - Posted: 21 Aug 2012, 23:26:17 UTC - in response to Message 1273784.

Why? And yes I did receive some for the Nvidia gpu during the last server hiccup. Did those too on the gpu. Run times were posted in an earlier post.

b/c a number of us haven't seen any VLARs crunch on our GPUs since the "fix" was put in place. in the Bug in server affecting older BOINC clients with NVIDIA GPUs thread, someone was asking troubleshooting questions a few days ago. after solving his problem, he made mention of some VLARs, which briefly led Erik Corpela to think that the VLARs were broken again. but it was quickly pointed out that his VLARs occurred on August 11th, before the fixes were put into place. so it seemed like a false alarm, but then you mentioned having VLARs on your nVidia GPU, so now i'm wondering again...

to be clear, i'm not having problems with this phenomenon myself.

____________

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,497,555
RAC: 25,398
United States
Message 1273795 - Posted: 21 Aug 2012, 23:43:47 UTC - in response to Message 1273789.

For anyone whom I might have confused, I have received
zero vlars for my Nvidia gpu since Eric's last fix on the
server bug. I will and do transfer some to my gpu to keep
it busy if I run out of regular work units and cannot get
any from the servers.
My apologies If I gave any other impression.

Profile SliverProject donor
Avatar
Send message
Joined: 18 May 11
Posts: 281
Credit: 7,162,769
RAC: 7,496
United States
Message 1273796 - Posted: 21 Aug 2012, 23:56:16 UTC

8/21/2012 7:50:44 PM | SETI@home | Started download of astropulse_6.04_opencl_nvidia_AUTHORS
8/21/2012 7:50:45 PM | SETI@home | Finished download of astropulse_6.04_opencl_nvidia_AUTHORS
8/21/2012 7:50:45 PM | SETI@home | Started download of astropulse_6.04_opencl_nvidia_COPYING
8/21/2012 7:50:48 PM | SETI@home | Finished download of astropulse_6.04_opencl_nvidia_COPYING
8/21/2012 7:50:48 PM | SETI@home | Started download of astropulse_6.04_opencl_nvidia_COPYRIGHT
8/21/2012 7:50:50 PM | SETI@home | Finished download of astropulse_6.04_opencl_nvidia_COPYRIGHT
8/21/2012 7:50:50 PM | SETI@home | Started download of astropulse_6.04_windows_intelx86__opencl_nvidia_100.pdb
8/21/2012 7:53:13 PM | SETI@home | Temporarily failed download of astropulse_6.04_windows_intelx86__opencl_nvidia_100.pdb: transient HTTP error
8/21/2012 7:53:13 PM | SETI@home | Backing off 3 min 33 sec on download of astropulse_6.04_windows_intelx86__opencl_nvidia_100.pdb
8/21/2012 7:53:13 PM | SETI@home | Started download of libfftw3f-3.dll


Ok, I'm stumped. What's with the nvidia AUTHORS/COPYING/COPYRIGHT, etc stuff? Also: What the heck is the last line there? (the .dll)
____________

Profile Fred E.Project donor
Volunteer tester
Send message
Joined: 22 Jul 99
Posts: 768
Credit: 24,139,004
RAC: 30
United States
Message 1273805 - Posted: 22 Aug 2012, 0:14:14 UTC
Last modified: 22 Aug 2012, 0:14:58 UTC

8/21/2012 7:50:44 PM | SETI@home | Started download of astropulse_6.04_opencl_nvidia_AUTHORS
8/21/2012 7:50:45 PM | SETI@home | Finished download of astropulse_6.04_opencl_nvidia_AUTHORS
8/21/2012 7:50:45 PM | SETI@home | Started download of astropulse_6.04_opencl_nvidia_COPYING
8/21/2012 7:50:48 PM | SETI@home | Finished download of astropulse_6.04_opencl_nvidia_COPYING
8/21/2012 7:50:48 PM | SETI@home | Started download of astropulse_6.04_opencl_nvidia_COPYRIGHT
8/21/2012 7:50:50 PM | SETI@home | Finished download of astropulse_6.04_opencl_nvidia_COPYRIGHT
8/21/2012 7:50:50 PM | SETI@home | Started download of astropulse_6.04_windows_intelx86__opencl_nvidia_100.pdb
8/21/2012 7:53:13 PM | SETI@home | Temporarily failed download of astropulse_6.04_windows_intelx86__opencl_nvidia_100.pdb: transient HTTP error
8/21/2012 7:53:13 PM | SETI@home | Backing off 3 min 33 sec on download of astropulse_6.04_windows_intelx86__opencl_nvidia_100.pdb
8/21/2012 7:53:13 PM | SETI@home | Started download of libfftw3f-3.dll

Ok, I'm stumped. What's with the nvidia AUTHORS/COPYING/COPYRIGHT, etc stuff? Also: What the heck is the last line there? (the .dll)


Looks like the new AP-for-Nvidia stock application was released today. You are getting all the files necessary to run it. That's a first-time only event for a new application. You are running stock, so you get it automatically if your prefs are set for "use Nvidia" and AP6.
____________
Another Fred
Support SETI@home when you search the Web with GoodSearch or shop online with GoodShop.

Profile SliverProject donor
Avatar
Send message
Joined: 18 May 11
Posts: 281
Credit: 7,162,769
RAC: 7,496
United States
Message 1273807 - Posted: 22 Aug 2012, 0:17:23 UTC - in response to Message 1273805.

Oh cool! Thanks for the help, bud :D
____________

Profile Keith Myers
Volunteer tester
Avatar
Send message
Joined: 29 Apr 01
Posts: 169
Credit: 63,507,267
RAC: 19,781
United States
Message 1273823 - Posted: 22 Aug 2012, 1:32:44 UTC - in response to Message 1273805.

So, how to get this new app for Nvidia OpenCL AstroPulse if you are running current Lunatics stuff on an anonymous platform?

Keith

____________

ClaggyProject donor
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4098
Credit: 33,057,138
RAC: 8,390
United Kingdom
Message 1273854 - Posted: 22 Aug 2012, 4:40:32 UTC - in response to Message 1273823.

So, how to get this new app for Nvidia OpenCL AstroPulse if you are running current Lunatics stuff on an anonymous platform?

Keith

Why would you want to change, you're already running r1316, Stock 6.04 is r1316 too.

Claggy

Profile Sunny129
Avatar
Send message
Joined: 7 Nov 00
Posts: 190
Credit: 3,163,755
RAC: 0
United States
Message 1273874 - Posted: 22 Aug 2012, 5:29:20 UTC - in response to Message 1273854.
Last modified: 22 Aug 2012, 6:21:48 UTC

So, how to get this new app for Nvidia OpenCL AstroPulse if you are running current Lunatics stuff on an anonymous platform?

Keith

Why would you want to change, you're already running r1316, Stock 6.04 is r1316 too.

Claggy

i'm not sure i understand what you're saying...the r1316 build didn't appear until 7/6/12, long after the most current Lunatics installer v0.40 was released (which included no OpenCL apps, let alone any apps on the r1316 build). if Keith is running the latest Lunatics installer, then he has an app_info.xml and is running anonymous platform. won't that prevent his host from automatically DLing OpenCL AP tasks for his nVidia GPU, even if APv6 are checked in his web preferences? doesn't this new OpenCL AP app have to be manually placed in the SETI@Home data directory, and isn't an entry required in the app_info.xml file in order to reference this new app and receive work for it?

Keith, see Raistmer's first post in THIS thread for a link to the Lunatics webpage that has the new OpenCL app available for download. that thread i linked you to also have a sample app_info.xml entry for nVidia GPUs in the 2nd post.
____________

Profile BilBg
Volunteer tester
Avatar
Send message
Joined: 27 May 07
Posts: 2692
Credit: 6,089,360
RAC: 4,554
Bulgaria
Message 1273886 - Posted: 22 Aug 2012, 6:51:02 UTC - in response to Message 1273874.


He already have the app, one of his tasks:
http://setiathome.berkeley.edu/result.php?resultid=2566706424

AstroPulse v.6
Non-graphics FFTW USE_CONVERSION_OPT
Windows x86 rev 1316, V6 match, by Raistmer with support of Lunatics.kwsn.net team. SSE2


Obviously he manually added it.


____________



- ALF - "Find out what you don't do well ..... then don't do it!" :)

Profile Keith Myers
Volunteer tester
Avatar
Send message
Joined: 29 Apr 01
Posts: 169
Credit: 63,507,267
RAC: 19,781
United States
Message 1273931 - Posted: 22 Aug 2012, 11:27:01 UTC - in response to Message 1273854.
Last modified: 22 Aug 2012, 11:30:04 UTC

I did not know that. I only saw the News announcement of new Nvidia OpenCL AstroPulse apps to be made available shortly after the ATI ones were sent out as stock apps. Would have been less confusing if someone, somewhere had mentioned that the stock 6.04 app was r1316.

Cheers, Keith
____________

ClaggyProject donor
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4098
Credit: 33,057,138
RAC: 8,390
United Kingdom
Message 1274109 - Posted: 22 Aug 2012, 20:46:25 UTC
Last modified: 22 Aug 2012, 20:46:39 UTC

The ucb Routers page, along with all the Cricket Graphs are down.

Claggy
____________


Profile [seti.international] Dirk SadowskiProject donor
Volunteer tester
Avatar
Send message
Joined: 6 Apr 07
Posts: 7086
Credit: 60,344,317
RAC: 15,639
Germany
Message 1274140 - Posted: 22 Aug 2012, 22:19:14 UTC
Last modified: 22 Aug 2012, 22:20:40 UTC

I would like to make one message because of the .vlar WUs.

I let run CUDA WUs since the project admins decided to send out a CUDA application.
Immediately I bought only because of SETI@home NVIDIA GPUs. ;-D

The project admins decided not to send .vlar WUs to NVIDIA GPUs (this are SETI@home Enhanced (MultiBeam) WUs with < 0.12 AR).
Because they slow down (other daily work on) the whole machine.

I think maybe because of: The member could think: "Hey, what's up?" - and will never again crunch SETI@home WUs.

The most members of SETI@home (maybe 95 %) let run the stock applications.

The stock 6.08_cuda up to 6.10_cuda_fermi applications calculate very bad/long this .vlar WUs.

Also on pre-Fermi GPUs the calculation is very slow.

I can't remember which application it was, maybe CUDA_V12 or x32f..
On my machine E7600 with GTX260 O/C:
A shorty: 2 minutes (120 seconds)
A normal 0.448x AR WU: 8 minutes and 20 seconds (500 seconds)
A 0.12x AR WU (which will be send): 25 minutes and 8 seconds (1508 seconds)
< 0.12 AR WU (.vlar WU, not send to NVIDIA GPUs): 1 hour and 30 minutes (5400 seconds)

This mean between normal (0.448x) and nearly VLAR WU (0.12x) the time is x ~ 3.

A .vlar WU would mean x ~ 10.8 longer than a normal AR WU. Wasted performance.


Maybe in future the project admins decide to send out .vlar WUs to Fermi and later GPUs - and still not to pre-Fermi GPUs (if this is possible with the server software).
If the upcoming stock S@h v7 CUDA application calculate very well this kind of WUs on Fermi+ GPUs.


* Best regards! :-) * Sutaru Tsureku, team seti.international founder. * Optimize your PC for higher RAC. * SETI@home needs your help. *
____________
BR

SETI@home Needs your Help ... $10 & U get a Star!

Team seti.international

Das Deutsche Cafe. The German Cafe.

ClaggyProject donor
Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 4098
Credit: 33,057,138
RAC: 8,390
United Kingdom
Message 1274146 - Posted: 22 Aug 2012, 22:34:36 UTC - in response to Message 1274109.
Last modified: 22 Aug 2012, 22:34:45 UTC

The Cricket Graphs are back, but are empty of fresh data,

Claggy
____________


Profile Wiggo
Avatar
Send message
Joined: 24 Jan 00
Posts: 6949
Credit: 94,621,878
RAC: 75,224
Australia
Message 1274154 - Posted: 22 Aug 2012, 23:19:39 UTC - in response to Message 1274140.
Last modified: 22 Aug 2012, 23:20:49 UTC

Maybe in future the project admins decide to send out .vlar WUs to Fermi and later GPUs - and still not to pre-Fermi GPUs (if this is possible with the server software).
If the upcoming stock S@h v7 CUDA application calculate very well this kind of WUs on Fermi+ GPUs.

Even though Fermi and later don't seem to have to many problems with them, their increased computation times (3-5 times longer than average) really throw the DCF out a lot, so I can't see the point in allowing them back ATM.

Cheers.
____________

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,497,555
RAC: 25,398
United States
Message 1274164 - Posted: 22 Aug 2012, 23:57:35 UTC - in response to Message 1274140.

A reply to some points made by Sutaru

I would like to make one message because of the .vlar WUs.

I let run CUDA WUs since the project admins decided to send out a CUDA application.
Immediately I bought only because of SETI@home NVIDIA GPUs. ;-D

The project admins decided not to send .vlar WUs to NVIDIA GPUs (this are SETI@home Enhanced (MultiBeam) WUs with < 0.12 AR).
Because they slow down (other daily work on) the whole machine.

bill- If this is a problem then yes, the offending work units should not be crunched on the gpu.

I think maybe because of: The member could think: "Hey, what's up?" - and will never again crunch SETI@home WUs.

bill - If those people are so easily dissuaded from crunching for SETI
without even investigating as to why this is happening and how to fix the problem then it's probably for the best that they don't do any Boinc project work at all because they will eventually run into problems with any project they choose.

The most members of SETI@home (maybe 95 %) let run the stock applications.

The stock 6.08_cuda up to 6.10_cuda_fermi applications calculate very bad/long this .vlar WUs.

bill - Yes the work units take longer. That is not bad unless pc sluggishness
or errors are a problem. Back in the beginning of the project all work units used to take a long time. The work still needs to be done though.

Also on pre-Fermi GPUs the calculation is very slow.

I can't remember which application it was, maybe CUDA_V12 or x32f..
On my machine E7600 with GTX260 O/C:
A shorty: 2 minutes (120 seconds)
A normal 0.448x AR WU: 8 minutes and 20 seconds (500 seconds)
A 0.12x AR WU (which will be send): 25 minutes and 8 seconds (1508 seconds)
< 0.12 AR WU (.vlar WU, not send to NVIDIA GPUs): 1 hour and 30 minutes (5400 seconds)

This mean between normal (0.448x) and nearly VLAR WU (0.12x) the time is x ~ 3.

A .vlar WU would mean x ~ 10.8 longer than a normal AR WU. Wasted performance.

bill - Can't agree. The vlar will take even longer to work on a cpu than a gpu.
To me crunching a work unit faster is better. If your gpu is not busy doing any other work units, why not let it do vlars if they cause no problems.

Maybe in future the project admins decide to send out .vlar WUs to Fermi and later GPUs - and still not to pre-Fermi GPUs (if this is possible with the server software).

bill - That would be excellent

If the upcoming stock S@h v7 CUDA application calculate very well this kind of WUs on Fermi+ GPUs.

bill - Hopefully. I have no objection to aborting vlars, or any other type of work unit, if they cause problems
on the client computer. Taking longer to run is not a problem in and of itself,
though.

* Best regards! :-) * Sutaru Tsureku, team seti.international founder. * Optimize your PC for higher RAC. * SETI@home needs your help. *

Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · 15 . . . 21 · Next

Message boards : Number crunching : Panic Mode On (76) Server Problems?

Copyright © 2014 University of California