Posts by Arvid Almstrom


log in
1) Message boards : Number crunching : How to contract BOINC develpoers (Message 1446877)
Posted 148 days ago by Profile Arvid Almstrom
In your email you don't say which version Boinc this is on, looking at your hosts at Einstein shows most of your hosts are running outdated Alpha versions of Boinc,
Boinc 7.2.5 was never a Public release, Make sure you're running one of the Recommended Releases, either 7.2.28, or 7.2.31, and if you're going to run Alpha releases please keep up to date,
the changes and links are all detailed in the BOINC 7.1/7.2 Change Log and News thread.

Claggy


I know I should upgrade and will do that. However, the way I found / verified the the problem via the source code for 'boinc_cmd.cpp' and 'gui_rpc_client.h'. There is a struct RPC_CLIENT which contains the int, which I think needs to be a long.


struct RPC_CLIENT {
int sock;
double start_time;
double timeout;

int set_global_prefs_override_struct(GLOBAL_PREFS&, GLOBAL_PREFS_MASK&);
int get_cc_config(CONFIG& config, LOG_FLAGS& log_flags);
int set_cc_config(CONFIG& config, LOG_FLAGS& log_flags);
int get_daily_xfer_history(DAILY_XFER_HISTORY&);
};


Hope to hear something back, but I guess it will take some time.

Thanks

Arvid
2) Message boards : Number crunching : How to contract BOINC develpoers (Message 1446763)
Posted 148 days ago by Profile Arvid Almstrom
Thanks for that Claggy, I'll give a try.

Arvid
3) Message boards : Number crunching : How to contract BOINC develpoers (Message 1446744)
Posted 148 days ago by Profile Arvid Almstrom
I wanted to ask the Boinc developers to change the daily_transfer from an int to a long as it has been wrapping around to -2GB for the last month.

My problem is that when I try to create and account for their message board, the page code that is to get the ReCaptha from Google is not displaying. Hence, I cannot create an account.

So does anyone here have contact with the people who manage the site for them so that they could fix the 'Create Account' page, or forward this on to someone in the developer team.

Many thanks

Arvid
4) Message boards : Number crunching : Help needed to enable running of multiple task on GPU (Message 1396421)
Posted 267 days ago by Profile Arvid Almstrom

Nice it works.

.50 for CPU on AP will reserve a core if you not use HT, that will make your cruncher a little less efective, who much i can´t tell, to find the only way is if you try both then choose the best one. On my 2x690 host i use 1 CPU and not use HT.

I allocate 1 CPU for each AP task because is the optimall setting in my 690 GPU´s i try 0.5 and works slower (the cpu can´t feed the GPU at full speed). Most of the posts about that on the forums indicates that is the best setting. My sugestion try both and choose the best one.

About the MB you don´t need to change the cpu time keep that low, makes little or non diference. Keep that at 0.06.

MB 0.24 GPU will make each one of your GPU running 1 AP and 2 MB (or 4 when no AP work where avaiable) I belive that could be very close to the optimal setting for you GPU´s


Thanks for that Juan, I'll run it for a while and see which option works best if I could only get some AP WU's to test with.

Arvid
5) Message boards : Number crunching : Help needed to enable running of multiple task on GPU (Message 1396411)
Posted 267 days ago by Profile Arvid Almstrom
@Arvid

First run the Lunnatics installer and enable MB cuda50 & GPU AP crunching then

Leave the app_info.xml as the lunnatics installer create it.

With Boinc 7.0.64 or + if you want to run more than one WU at a time the better way to do that is by create a new file who´s name is: app_config.xml

This is a working app_config.xml

<app_config>
<app>
<name>setiathome_v7</name>
<gpu_versions>
<gpu_usage>0.49</gpu_usage>
<cpu_usage>0.06</cpu_usage>
</gpu_versions>
</app>
<app>
<name>setiathome_enhanced</name>
<gpu_versions>
<gpu_usage>0.49</gpu_usage>
<cpu_usage>0.06</cpu_usage>
</gpu_versions>
</app>
<app>
<name>astropulse_v6</name>
<gpu_versions>
<gpu_usage>0.51</gpu_usage>
<cpu_usage>1.0</cpu_usage>
</gpu_versions>
</app>
</app_config>

This file allow you to run 1 AP task on each GPU and 1 MB at a time (or 2 MB if no AP WU are avaiable), put it to work first after you sure it´s working you could easely change to run more without even need to reboot the BOinc itself (just by reading the configuration file again). But remember EACH AP task need a Free core to work properly, so keep that in mind when you make your configuration. Even if each one of your GPU can easely run 4 AP, that will uses all the cores on a I7 if HT is off (the optimal setting for SETI crunching).

Other important thing is, if you create the file with the notepath take care that you don't save with the extension .txt (the normal notepath extension) or the file will not work.


Juan, thanks for the info.

I opted for .51 gpu and .5 cpu for the AP.

I then selected .24 gpu and .16 CPU for the MB.

I wanted 1 AP and 2 MB tasks at once, but with the AP at .51 I had to go for 1 AP and 2 MB, or 4 x MB, but I have 3GB RAM on the card and it is the new shiny 780, so it should cope with it I hope.

I will see how it goes if I ever get any AP's from the server.

Why have you allocated a full CPU for the AP task? Does it really need a full CPU process?

Many thanks

Arvid
6) Message boards : Number crunching : Help needed to enable running of multiple task on GPU (Message 1396408)
Posted 267 days ago by Profile Arvid Almstrom


In the past I just installed the optimized apps, which created the correct file and it was just a matter of doing a search and replace for the <count>, but I can't install the optimized apps with version 7.



Why can't you install the opti apps? The Lunatics installer should run just fine on your rig.


Gatekeeper, I thought as I was getting errors from the previously installed Opt Apps, and the issue with Lunatics license issue, that it wouldn't work.

However, thank you for this as there is a new version as well as the license issue must have been resolved and they are allowed to publish the software.

Arvid
7) Message boards : Number crunching : Help needed to enable running of multiple task on GPU (Message 1396287)
Posted 267 days ago by Profile Arvid Almstrom
I have just installed 2 x GTX 780 into my computer and re-enabled S@H on this computer, as E@H does not run well on the new nVidia cards (only CUDA 3.2).

I got error messages about unsupported app in xlm file, so I reset the project.

After it downloaded the app and WU's it started to run, but only runs with 1 WU per GPU.

I tried to create the missing app_info.xml file from various posts and wiki's


<app_info>
<app_config>
<app>
<name>astropulse_v6</name>
<max_concurrent>14</max_concurrent>
<gpu_versions>
<gpu_usage>0.25</gpu_usage>
<cpu_usage>0.12</cpu_usage>
</gpu_versions>
</app>
<app>
<name>setiathome_enhanced_v7</name>
<max_concurrent>14</max_concurrent>
<gpu_versions>
<gpu_usage>0.25</gpu_usage>
<cpu_usage>0.12</cpu_usage>
</gpu_versions>
</app>
<app>
<name>setiathome_enhanced</name>
<max_concurrent>14</max_concurrent>
<gpu_versions>
<gpu_usage>0.25</gpu_usage>
<cpu_usage>0.12</cpu_usage>
</gpu_versions>
</app>
</app_config>
</app_info>


However, this causes the scheduler to not issue tasks as [work_fetch] informs me that there is not app.

I guess I have to fill in the file_info flags, which I can now do for the V7 CUDA files, but I don't have the other applications, nor do I have all the other required info for the file.

Is there a default app_info.xml file that I can get? Why is this so hard to do, it should be easier, like a setting in 'Computing Preferences'.

In the past I just installed the optimized apps, which created the correct file and it was just a matter of doing a search and replace for the <count>, but I can't install the optimized apps with version 7.

Any help would be most appreciated.

Arvid
8) Message boards : Number crunching : Which card is best? (Message 1301420)
Posted 536 days ago by Profile Arvid Almstrom
Hi Nicholas

If you are after nVidia cards, why don't you start here:

http://www.nvidia.com/content/HelpMeChoose/fx2/HelpMeChoose.asp?lang=en-us

I would probably look at the 650 Ti based on processing power, but check the boards of any issues as this is a brand new card.

NewEgg have plenty of these right in your price target.

Arvid
9) Message boards : Number crunching : CLOSED* SETI/BOINC Milestones [ v2.0 ] - XXIV *CLOSED (Message 1292589)
Posted 562 days ago by Profile Arvid Almstrom
Way to go Arvid !!!


Thanks Spyros, and how did you go with the updated GPU logo, did anyone manage to find and update it?

Arvid
10) Message boards : Number crunching : CLOSED* SETI/BOINC Milestones [ v2.0 ] - XXIV *CLOSED (Message 1292564)
Posted 562 days ago by Profile Arvid Almstrom
Just broke though the 100M mark or should I say;

100,000,000 to make it look bigger.

This however, took a little longer as my main machine is still not quite happy after running out of WU's, but it is on the mend.

Arvid
11) Message boards : Number crunching : new WU numbering scheme (Message 1291095)
Posted 566 days ago by Profile Arvid Almstrom
I have just noticed that these new WU's take a lot longer to process on the GPU.

The WU's on my computer states that the expected time is around 13 minutes to complete, but it takes me around 30 minutes for the GTX670, running 4 tasks per GPU, to complete the task.

I first thought that the new nVidia drivers could be causing problems, but looking that old fashioned WU's they process at the normal rate that it states. I just hope that these are not some form of VLAR WU's that are pretending to be normal tasks.

Has anyone else notices the run-time difference, running them on the CPU looks to run at the estimated time.

Arvid
12) Message boards : Number crunching : Panic Mode On (77) Server Problems? (Message 1290475)
Posted 567 days ago by Profile Arvid Almstrom
I have just had a record of WU's being processed on my computer, 6 out of 8 slots were processing. Unfortunately, it only lasted about 1 1/2 minute and I am now back to 0 of 8.

I think the uploading / reporting and general slowness is related to the AP Splitters running and that this or their WU's somehow causes the project to take a major dive.

I have not had WU's for my main computer for many days and now I have received a few tasks but I cannot report them or my status to ask for more work.

I hope something gets discovered soon, and I hope that the staff are aware of the problems we are having with communication to the SETI project as a whole.

Arvid
13) Message boards : Number crunching : Panic Mode On (77) Server Problems? (Message 1290470)
Posted 567 days ago by Profile Arvid Almstrom
What is the explanation behind a "shorties storm" ? They don't seem to originate from the same tapes? Yet almost anything being sent out are VHAR units. Is this a server problem? I'm curious to read some info on this ;) Works flows slowly at best right now, better than nothing at all of course.

Kind regards.

In simple terms: SETI gets its data for free, by taking its own copy of the data being recorded during the course of astronmonical observations at the Arecibo radio telescope.

Different groups of radio astronomers are allocated observing time on the telescope, according to an observatory schedule which can be searched online if you're really interested. Each separate group of observers has control of the telescope during their assigned time slot, and control its movement and observing patterns.

Some astronomers are interested in long, steady, deep-space observations of, near enough, point sources. The focal point of the telescope remains steady in relation to the sky - the recordings have a low 'angle range' between the beginning and end of the 109 seconds we study in each workunit. Those sessions create the 'VLAR' tasks when we get to crunch the recordings.

Other observing teams are more interested in fast surveys of large parts of the sky. They use the observatory's radio antenna in what is known as a 'basketweave' mode, with the telescope nodding from side to side while the earth turns under the sky. That leads to the high angle range tasks - we know them as 'shorties', because it's not worth doing such intense analysis when potential signal sources remain in the field of view for such a short time.

And in between, there are observations - or even recordings taken during telescope maintenance - where the antenna is not being actively steered at all, but simply receiving whatever happens to be coming from the sky patch directly overhead as the earth turns. That gives us the normal, mid-AR tasks which form our staple diet.


Hi Richard,

Thank you for this information. I have been crunching SETI work for 12 1/5 years and I did not know this fact. I thought it had to do with the angle that the antenna was pointing at, relative to the sky. and that this has some issues with not picking up the same amount of information and was degraded for some reason or other.

Many thanks for this info,

Arvid
14) Message boards : Technical News : Hot Day (Oct 02 2012) (Message 1290466)
Posted 567 days ago by Profile Arvid Almstrom
Hi Matt

Thanks for the update on the current state of things and good luck trying to keeping cool.

I was interested if you had any thoughts on the network problems of the past 7-10 days. From a novice, it looks like there was, over this last week, a correlation between AP splitters running and the whole SETI project stopping to respond.

Many thanks for the update and keep 'em coming.

Arvid
15) Message boards : Number crunching : GTX 690 conundrum - only Seti Multibeam don't run on GPUs (Message 1289009)
Posted 571 days ago by Profile Arvid Almstrom
I have read many threads where people have been trying to get 4 x 690's to work and have not succeeded due to hardware resource limits.

In one thread, cant remember who it was sorry, he had been getting custom firmware from nVidia / card manufacturer to try an overcome this problem, but that only killed his card.

Have you tried to remove one of the cards and see if 3 x 690's would work, it would then save you from re-installing Windows.

Just a thought.

Arvid
16) Message boards : Number crunching : Unable to install on new computer (Message 1288335)
Posted 573 days ago by Profile Arvid Almstrom
I have now been able to get a few WU's. It was just strange that the download, which normally showes the binaries, didn't happen.

However, the optimized app is now running, let just hope the server starts acting normal soon and gives out some more WU's.

Arvid
17) Message boards : Number crunching : Unable to install on new computer (Message 1288313)
Posted 573 days ago by Profile Arvid Almstrom
I just tried to install SETI on a new computer but it doesn't download the apps.

All I get are the 4 image files and nothing else. I have tried resetting and removing SETI but still I only get the 4 image files.

Is there anything I can try or is this a server issue?

EDIT: I have installed Lunatics apps but every time I try and download, the server response is Scheduler request completed: got 0 new tasks


Arvid
18) Message boards : Number crunching : Optimize your GPU. Find the value the easy way. (Message 1280069)
Posted 595 days ago by Profile Arvid Almstrom
Here are my tests running on a 570 with different CUDA apps.

For some reason running 4 tasks didn't complete on either the 41z cuda 3.2 or 41z cuda 4.2.

Starting automatic test: (x41x_winx64_cuda41)
Device: 0, device count: 1, average time / count: 127, average time on device: 127 Seconds (2 Minutes, 7 Seconds)
Device: 0, device count: 2, average time / count: 203, average time on device: 101 Seconds (1 Minutes, 41 Seconds)
Device: 0, device count: 3, average time / count: 290, average time on device: 96 Seconds (1 Minutes, 36 Seconds)

Starting automatic test: (x41x_winx64_cuda42)
Device: 0, device count: 1, average time / count: 124, average time on device: 124 Seconds (2 Minutes, 4 Seconds)
Device: 0, device count: 2, average time / count: 198, average time on device: 99 Seconds (1 Minutes, 39 Seconds)
Device: 0, device count: 3, average time / count: 285, average time on device: 95 Seconds (1 Minutes, 35 Seconds)
Device: 0, device count: 4, average time / count: 469, average time on device: 117 Seconds (1 Minutes, 57 Seconds)

Starting automatic test: (x41z_winx64_cuda42)
Device: 0, device count: 1, average time / count: 128, average time on device: 128 Seconds (2 Minutes, 8 Seconds)
Device: 0, device count: 2, average time / count: 198, average time on device: 99 Seconds (1 Minutes, 39 Seconds)
Device: 0, device count: 3, average time / count: 284, average time on device: 94 Seconds (1 Minutes, 34 Seconds)

Arvid
19) Message boards : Number crunching : Graphic card problems (Message 1243068)
Posted 684 days ago by Profile Arvid Almstrom
Unfortunately, EVGA wont warranty the cards if you arent the original owner....

The frequent crashes should have been the trigger that alerted you to the possibility of video card problems. Outside of crappy drivers, high temperatures are usually the cause of these crashes.

Have you tried booting these cards up from another computer (for just a few seconds) to determine if they are really dead?

If they don't come up then they are very dead and not worth reviving.

No other computer exists here, I'd thought of putting them up for sale as spare parts. I might though give one of them 1 one chance, just for curiosity, clean out any final dust and change the the thermal compound to AS5 on the gpus and the nf200 chip, if nothing happens then it's no loss as it won't cost Me any money to do so, just time and a little effort, which isn't all that much to do really, if it comes up then try on the 2nd as I'd read that the gpus just shut down over a certain temp to protect themselves.

If the cards won't boot up with both GPUs cold, changing the HS compound and cleaning them is not gonna do any good, as they have not even come up to temp at that point.

Only other outside chance of saving them is the infamous oven trick, stripping them down and baking the circuit boards in an effort to reflow the solder. This has met with mixed success....there are numerous posts and youtube clips on the subject.



Thank you for this suggestion. At first I thought it was a joke but thought I have nothing to loose.

I had a look at some videos and thought why not. I had a GTX 280 in the bin that I swapped out for a client. The card would work with 2D but as soon as you stressed the card it would BSOD.

I followed suggestions, Temp 190C for 9 minutes, let it cool down, reassembled it and presto. It has been working for the last hour doing Seti, I hope it keeps going.

Arvid


18-hours and still going strong after the baking and no errors.


http://setiathome.berkeley.edu/results.php?hostid=6685901

Arvid
20) Message boards : Number crunching : Graphic card problems (Message 1242649)
Posted 685 days ago by Profile Arvid Almstrom
Unfortunately, EVGA wont warranty the cards if you arent the original owner....

The frequent crashes should have been the trigger that alerted you to the possibility of video card problems. Outside of crappy drivers, high temperatures are usually the cause of these crashes.

Have you tried booting these cards up from another computer (for just a few seconds) to determine if they are really dead?

If they don't come up then they are very dead and not worth reviving.

No other computer exists here, I'd thought of putting them up for sale as spare parts. I might though give one of them 1 one chance, just for curiosity, clean out any final dust and change the the thermal compound to AS5 on the gpus and the nf200 chip, if nothing happens then it's no loss as it won't cost Me any money to do so, just time and a little effort, which isn't all that much to do really, if it comes up then try on the 2nd as I'd read that the gpus just shut down over a certain temp to protect themselves.

If the cards won't boot up with both GPUs cold, changing the HS compound and cleaning them is not gonna do any good, as they have not even come up to temp at that point.

Only other outside chance of saving them is the infamous oven trick, stripping them down and baking the circuit boards in an effort to reflow the solder. This has met with mixed success....there are numerous posts and youtube clips on the subject.



Thank you for this suggestion. At first I thought it was a joke but thought I have nothing to loose.

I had a look at some videos and thought why not. I had a GTX 280 in the bin that I swapped out for a client. The card would work with 2D but as soon as you stressed the card it would BSOD.

I followed suggestions, Temp 190C for 9 minutes, let it cool down, reassembled it and presto. It has been working for the last hour doing Seti, I hope it keeps going.

Arvid


Next 20

Copyright © 2014 University of California