Posts by petri33

1) Message boards : Number crunching : Linux CUDA 'Special' App finally available, featuring Low CPU use (Message 1913630)
Posted 5 hours ago by Profile petri33Project Donor
Post:
Well, I don't know what Petri's doing with that zi3xs4 version, but it sure doesn't look stable.

Workunit 2813410758 (blc05_2bit_guppi_57976_77315_HIP46417_0038.12216.818.21.44.188.vlar)
Task 6306343836 (S=0, A=0, P=30, T=0, G=0, BS=13.40984, BG=0, BP=23.54865) x41p_zi3xs4, Cuda 9.10 special
Task 6306343837 (S=2, A=2, P=5, T=2, G=0, BS=24.3649, BG=0, BP=12.14392) x41p_zi3v, Cuda 9.00 special

Workunit 2813410770 (blc05_2bit_guppi_57976_75329_HIP46343_0032.11400.818.21.44.192.vlar)
Task 6306343840 (S=0, A=0, P=30, T=0, G=0, BS=12.58375, BG=0, BP=2.555692) x41p_zi3xs4, Cuda 9.10 special
Task 6306343841 (S=21, A=0, P=5, T=0, G=0, BS=24.6841, BG=0, BP=9.582356) x41p_zi3v, Cuda 9.00 special

Workunit 2813438980 (blc05_2bit_guppi_57976_76984_HIP46432_0037.16675.409.21.44.90.vlar)
Task 6306402535 (S=0, A=0, P=30, T=0, G=0, BS=12.35322, BG=0, BP=3.181179) x41p_zi3xs4, Cuda 9.10 special
Task 6306402536 (S=0, A=0, P=8, T=1, G=0, BS=23.56962, BG=0, BP=0.9703487) x41p_zi3v, Cuda 8.00 special

He's coughing up 30-Pulse hairballs where my zi3v hosts are reporting normal-looking results. And, of the Pulses that are reported by my hosts, I don't see any correlation with his reported Pulses.


I hope someone has those wu's saved somewhere. They might reveal a bug in off line testing.
I tried the links and the wu is not available any more and the one in parentheses gives a ngnix error).
2) Message boards : Number crunching : Panic Mode On (109) Server Problems? (Message 1913261)
Posted 2 days ago by Profile petri33Project Donor
Post:
My backup project is to improve the code and run local tests.
3) Message boards : Number crunching : Panic Mode On (109) Server Problems? (Message 1913240)
Posted 2 days ago by Profile petri33Project Donor
Post:
Results ready to send : one ????
EDIT: Just changed to zero.
4) Message boards : Cafe SETI : What Are You Reading? (Message 1908973)
Posted 23 days ago by Profile petri33Project Donor
Post:
Source code.
I'm reading the source code.
The source code written in C++ and CUDA.
I do not know why, but it is rather captivating.

Petri
5) Message boards : Cafe SETI : Word Link # 94 (Message 1908972)
Posted 23 days ago by Profile petri33Project Donor
Post:
Bells


Whiskey
6) Message boards : Cafe SETI : X2017 chrunching environments and surroundings (Message 1908971)
Posted 23 days ago by Profile petri33Project Donor
Post:
You are welcome.


<3

Thx Uli!
7) Message boards : Number crunching : Linux CUDA 'Special' App finally available, featuring Low CPU use (Message 1908970)
Posted 23 days ago by Profile petri33Project Donor
Post:
Two solutions come to mind:
No "Anne Onee Moose" accounts.
Second automatically stop sending tasks to computers with high invalid rates in the same way that high computing error rate computers get reduced task allowances.


I like the word rate (not count) as in ratio compared to the number of good ones or RAC.
A totally valid host can get into an error mode sometimes. It should have a possibility to recover rather hastly.
A good computer can have a bad behaving new (a brand new beta) version of an executable but as soon as it has been fixed it should be allowed to recover.

Those new computers with RAC < xx and valid tasks < yy could get spanked, ignored, rectified or coerced to use more suitable running environments (less dust, right drivers, ...) , settings (no OC or OV) or correct versions(up to par and date) of drivers and software.

I know my host can be seen/taken as one of the worst pollutants/misbehaving ones but I have a cause and an intent to make it (and the whole computing) better.

And a good night to all of you at 3 AM here.

Petri
8) Message boards : Cafe SETI : X2017 chrunching environments and surroundings (Message 1908643)
Posted 25 days ago by Profile petri33Project Donor
Post:
https://drive.google.com/open?id=1wK6aInEptzeErxuTNXaRfez2je5TyLVG

Petri

p.s. thank you for those lovely Christmas cards
9) Message boards : Number crunching : Seti@Home or Atropulse? (Message 1908202)
Posted 28 days ago by Profile petri33Project Donor
Post:
... and the MB credit is highly dependent on the angle rate (ar) of the packet. So is the runtime between different apps.
http://setiathome.berkeley.edu/workunit.php?wuid=2784838125
MB can pay more than one credit per second. I have not seen AP do that.

Petri
10) Message boards : Number crunching : Panic Mode On (108) Server Problems? (Message 1905280)
Posted 7 Dec 2017 by Profile petri33Project Donor
Post:
Manually releasing ghosts 20 at a time. Yeah.
Give me a release ghosts button.
It's not that bad :) It recovers 20 if it can. It releases what it can't recover i.e. old ones.

EDIT: I have wondered how you ghost so many ... is it because of corrupt client_state because of the large amount of tasks it is trying to handle?


Sometimes one of the GPUs goes to an error state and all started apps begin to say 'can not determine number of CPUs' and the tasks error out. If I hit reset project then before they are uploaded they become ghosts. That is my explanation.
11) Message boards : Number crunching : Panic Mode On (108) Server Problems? (Message 1905212)
Posted 6 Dec 2017 by Profile petri33Project Donor
Post:
Wed 06 Dec 2017 10:52:17 PM EET | SETI@home | Project has no tasks available
Seem like none have been splut to my machine. (split, splat splut :) )
If you let loose your pile of ghosted tasks you would have a better chance of getting more :D


Manually releasing ghosts 20 at a time. Yeah.
Give me a release ghosts button.
12) Message boards : Number crunching : Panic Mode On (108) Server Problems? (Message 1905178)
Posted 6 Dec 2017 by Profile petri33Project Donor
Post:
Wed 06 Dec 2017 10:52:17 PM EET | SETI@home | Project has no tasks available

Seem like none have been splut to my machine. (split, splat splut :) )
13) Message boards : Number crunching : Panic Mode On (108) Server Problems? (Message 1900306)
Posted 10 Nov 2017 by Profile petri33Project Donor
Post:
Just run out of my CPU cache. The GPU cache was dry when I got home from work several hours ago...
[EDIT]
... and 600 000+ ready to send. Could we have some Arecibo vlars too. Pleeeeease.
14) Message boards : Number crunching : Linux CUDA 'Special' App finally available, featuring Low CPU use (Message 1899046)
Posted 4 Nov 2017 by Profile petri33Project Donor
Post:

p.s. What is a command line option -spike_fft_limit 4096 or similar (can not remember it right now) in SoG?

Petri


It shifts threshold for switching between 2 Spike computation strategies. One computes whole spike on single thread (so, 1D grid), another uses reduction and distributes computation over few workitems (threads) so 2D grid (with overheadon reduction though) so, for some matrix geometry one kernel better, for some - another. And this option allows user to move threshold for switching between them.


Thank you for the explanation.

Petri
15) Message boards : Number crunching : Linux CUDA 'Special' App finally available, featuring Low CPU use (Message 1898604)
Posted 1 Nov 2017 by Profile petri33Project Donor
Post:
Well, Petri says it's because his newer Apps are finding signals in the first chirp whereas the other Apps aren't. It is something in the newer Apps, he's just not sure what.
Other than that, the full zi3xs3 runs nicely on the Pascal GPUs.

To be correct: "first chirp" is zero chirp. And definitely algorithm looks for signals there (it means no relative motion regarding source and receiver).
What is omitted and by the reason is the 0th slot in PoT analysis (for all chirps). Zero slot means static signal strength and obviously should be ignored.
If Petri's app really accepts anything from that slot it's serious bug.
EDIT: indeed, handling 0th slot differently from all others means divergence and performance drop in CU that processed it along with others. But that's life, correct algorithm functioning requires omitting results from that slot.
If I recall correctly I implemented it in way that all processing is performed w/o deviation but results reduction omits anything from that slot. In such way GPU performance drop is minimal.


Hi,

Just like Raistmer said: Zero chirp is the first one and then the +- something ones. The fft PoT slot 0 for every chirp is the static (0 Hz) value and that is not used, it is omitted.

Divergence to a short path v.s. some other things: Any output value/value to be checked in the middle for action can be multiplied with factor = (PoT == 0 ? 0.0f : 1.0f); . One multiplication vs divergence to a path length zero can have an impact and it's performance can wary between the CUDA GPU generations/models. Current implementation prefers if(pot == 0) return; . That causes divergence (BAD thing for a GPU). Things may change, but pot 0 for any fft will never be in the reported signals. Chirp 0 will be checked as will all other chirps too.

p.s. What is a command line option -spike_fft_limit 4096 or similar (can not remember it right now) in SoG?

Petri
16) Message boards : Number crunching : Linux CUDA 'Special' App finally available, featuring Low CPU use (Message 1895161)
Posted 13 Oct 2017 by Profile petri33Project Donor
Post:
To find the best that is not reported is time consuming in a parallel world.

Yes. indeed. But still it's part of algorithm.
Until algorithm will be changed best should be found correctly (some sort of reduction from best per CU to single best could be used to reduce slowdown from serialization.
Regarding overflows - yep, early versions of SoG had same issue too. The more distributed task computation is the bigger amount of signals one should store to properly reorder on reporting. At some point it will too costly indeed. But if you still doing one icfft per kernel call amount of signals to keep should be not too huge.


+1, from the far end of the visible world. I'll be back.
17) Message boards : Number crunching : Invalid Host Messaging (Message 1895139)
Posted 13 Oct 2017 by Profile petri33Project Donor
Post:
I'm invalid for a week.

I'm a host to a parasite called quantum mechanics. The symptoms can be found with AI entering the Seti computation world. Nothing can be cured but by the acceptance. That is how the world works.

*No animals were hurt when writing this, I was not not hurt either, nor any religion is hurt or sponsored here for what|so|ever ... and you have all rights to your own feelings and expressions of them.* -- I'm OFF
18) Message boards : Number crunching : Linux CUDA 'Special' App finally available, featuring Low CPU use (Message 1895131)
Posted 13 Oct 2017 by Profile petri33Project Donor
Post:
My comment wasn't intended to suggest that the problem didn't need to be fixed. I was just noting that I've seen it appear with the stock Cuda apps, in particular the Cuda50 running on my daily driver. And I was passing along your analysis that it wasn't just a processing order issue.

Ideally, Jason would probably be the one to try to track it down in the current Cuda codebase, but he has been absent for awhile, so if it can be fixed in the Special App, I would expect that it could be ported back to the more widely used Cuda apps.

As I think I've expressed multiple times previously, just because a WU overflows doesn't mean that it's worthless. That 30 signal cutoff was based on storage considerations, not the value of the scientific data. The apps need to report consistently and let the scientists sort through the results and make any "noise bomb" determination. Anyway, what appears to be a noise bomb to one person might actually turn out to be an alien ABBA concert to another. ;^)


ABBA! Yes. One night in a "pulse nightmare" (an old one "jungle") no.

Choose your game. There are rules. Some hands in a Poker game give you points, some Ace in a sleeve does not. Change the rules - lower the limit to regard a 'signal' that is below the noise lever (i.e. below one scorewise) to be accepted. I could report all of them! To find the best that is not reported is time consuming in a parallel world. Be prepared for the quantum era.

Mamma Mia! No, no. The Winner does not take it all - it is just on the wings of the next hurricane named after a man - Fernando. - ABBA quote off.

Still no name until I fall asleep.
19) Message boards : Number crunching : Help me Optimize my GPUs (Message 1895116)
Posted 13 Oct 2017 by Profile petri33Project Donor
Post:
[2] NVIDIA GeForce GTX 980 (4095MB) driver: 384.76 OpenCL: 1.2
I need some help getting better results from them...


1) Install Linux.
2) get a Soft ware to do what a person has to do. (Missspellintentional)

--
Me

EDIT: if you want to triple your RAC.
20) Message boards : Number crunching : 16 GPUs? For real? (Message 1895115)
Posted 13 Oct 2017 by Profile petri33Project Donor
Post:
Having BOINC report more GPUs than exist in the system is also pretty easy & doesn't require modifying any code. Just tweaking some configs.


I tried that when I found that out from the source code, however it did not work for me. That is why I had to make some modifications to the boinc client code and I wanted make it to be visible for everyone and that is why the Ti is called Tu.

I'm still a week off. (browsing through the threads and vanishing -- screaming as I distance from the real world for a week off from duty <3)


Next 20


 
©2018 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.