SETI@Home Wow!-Event 2018

Page 5 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

crashtech

Lifer
Jan 4, 2013
10,578
2,146
146
Average for 1080Ti, X99, Linux, cuda90 application:
175 W, 89 % shader utilization, 67 % memory interface utilization
This includes the gaps between task teardown and bringup...<snip>
What I should have said was that I see high utilization, but not high temps, so whatever is going on in there doesn't seem to be taxing the GPU hardware as much as some other apps do. For instance, on my 7970's, I have to undervolt/underclock to run Milkyway (3 concurrent) without overheating during these times of high ambient temps. SETI presents no such difficulty, running below expected temps at max clocks on all GPUs, mostly with two concurrent.
 

StefanR5R

Elite Member
Dec 10, 2016
5,907
8,801
136
What I should have said was that I see high utilization, but not high temps, so whatever is going on in there doesn't seem to be taxing the GPU hardware as much as some other apps do.
That's true. The 1080Ti could pull 250 W, but it is content with 175 W for SETI cuda90 average.
Peaks: 207 W, 100 % sm, 93 % mem

PS,
here are the top 16 teams ordered by their production between 18:00 and 20:00 UTC.
Code:
   1    GPU Users Group         657,928
   2    SETI.Germany            253,822
   4    Gridcoin                107,741
   9    Planet 3DNow!            99,007
   3    [H]ard|OCP               96,264
   6    TeAm AnandTech           88,031
   7    ngb:setiteam             68,530
   8    Raccoon Lovers           58,282
  11    BOINC@AUSTRALIA          48,059
  10    L'Alliance Francophone   47,968
  13    USA                      45,371
  12    Czech National Team      43,180
  14    MVPs                     43,082
  15    The Knights Who Say Ni!  38,934
  16    Rechenkraft.net          31,284
   5    BOINC.Italy              22,791

PPS,
you haven't signed up for the event yet? You still can!
https://www.seti-germany.de/Wow/anmeldung.php?&lang=en
(The SETI user ID which you need to enter can be taken from your account page http://setiathome.berkeley.edu/home.php, top left table.)
 
Last edited:

crashtech

Lifer
Jan 4, 2013
10,578
2,146
146
That's true. The 1080Ti could pull 250 W, but it is content with 175 W for SETI cuda90 average.
Peaks: 207 W, 100 % sm, 93 % mem

...

PPS,
you haven't signed up for the event yet? You still can!
https://www.seti-germany.de/Wow/anmeldung.php?&lang=en
(The SETI user ID which you need to enter can be taken from your account page http://setiathome.berkeley.edu/home.php, top left table.)

Is that for me? I think I am signed up, currently #37 as of this moment:

https://www.seti-germany.de/Wow/stats_user_all.php?year=&lang=en
 

Howdy

Senior member
Nov 12, 2017
572
480
136
21/76 active users signed up for the TeAm!!!! We'd like the other 55 to sign up too (if they like) Call your friends, call your neighbors, call E.T, call anybody and everybody!!! Lets make it rain TeAm Seti points in here and find a green guy!!
 
Last edited:

ao_ika_red

Golden Member
Aug 11, 2016
1,679
715
136
21/76 active users signed up for the TeAm!!!! We'd like the other 55 to sign up too (if they like) Call your friends, call your neighbors, call E.T, call anybody and everybody!!! Lets make it rain TeAm Seti points in here and find a green guy!!
With only 2 new registered Wow! participants we already passed ngb:setiteam and will likely beat BOINC.Italy. Imagine what we'll get with 20 more.
 
Last edited:
Reactions: Howdy

StefanR5R

Elite Member
Dec 10, 2016
5,907
8,801
136
So far, I have 0 invalid tasks, but 64 inconclusive.
Clicked through some of my wingmen of these inconclusive tasks, and find gems like this:
State: All (2579) · In progress (6) · Validation pending (972) · Validation inconclusive (979) · Valid (0) · Invalid (622) · Error (0)
 

crashtech

Lifer
Jan 4, 2013
10,578
2,146
146
What the heck! I think I had a problem with one of my Ryzens for a while, seems either the PSU or PCI-e slot is going bad and making the GPU turn off suddenly at random. So I do have some errors, but no invalids.
 

lane42

Diamond Member
Sep 3, 2000
5,721
624
126
I could be wrong, but it looks like someone trying to setup cuda 9.2 on a linux machine.
That's the only thing I know of that gives that many validation inconclusive.
 

ao_ika_red

Golden Member
Aug 11, 2016
1,679
715
136
Clicked through some of my wingmen of these inconclusive tasks, and find gems like this:
State: All (2579) · In progress (6) · Validation pending (972) · Validation inconclusive (979) · Valid (0) · Invalid (622) · Error (0)
I could be wrong, but it looks like someone trying to setup cuda 9.2 on a linux machine.
That's the only thing I know of that gives that many validation inconclusive.
I'm still concerned with the invalid tasks though, because that's a lot.
 

lane42

Diamond Member
Sep 3, 2000
5,721
624
126
I think they have something to do with just bad workunits, i.e. very high angle range's.
An average is 0.4, looking around I see guy's with a lot of invalids with angle range's of
50.xxx, 100.xxx, 200.xxx. have never seen a/r of 200. I bet that guy's times with all the invalid's
is between 1 and 3 seconds.The big question is how did he get so many.
 

StefanR5R

Elite Member
Dec 10, 2016
5,907
8,801
136
The central issue with that machine isn't the high rate of invalids, it's the fact that it has 0 valid tasks. I bet none of its pendings and inconclusives will ever be validated. I hope the user goes to look at his results eventually, and takes it off SETI. (It's stock applications on old hardware. I think it was this host which I saw yesterday. It's now wingman of several of my inconclusive tasks.)

Another frequent source of invalid tasks are Mac OS X hosts with stock application. Those that I saw have many inclonclusive tasks, but at least also a few valid ones.
 

Kiska

Golden Member
Apr 4, 2012
1,037
291
136
Another frequent source of invalid tasks are Mac OS X hosts with stock application. Those that I saw have many inclonclusive tasks, but at least also a few valid ones.

Another frequent source is, running the latest Nvidia driver on Fermi or older with any OS
 
Reactions: ao_ika_red

ao_ika_red

Golden Member
Aug 11, 2016
1,679
715
136
The central issue with that machine isn't the high rate of invalids, it's the fact that it has 0 valid tasks. I bet none of its pendings and inconclusives will ever be validated. I hope the user goes to look at his results eventually, and takes it off SETI. (It's stock applications on old hardware. I think it was this host which I saw yesterday. It's now wingman of several of my inconclusive tasks.)

Another frequent source of invalid tasks are Mac OS X hosts with stock application. Those that I saw have many inclonclusive tasks, but at least also a few valid ones.
Holy cow. That's old. I don't even think Tesla is still supported by stock app.
 

StefanR5R

Elite Member
Dec 10, 2016
5,907
8,801
136
So I had put watt meters in front of a GPU host and of one of my best CPU hosts, in order to determine their respective power efficiency.

CPU host: dual Broadwell-EP
2x E5-2696 v4, 2x 22C/44T @ 2.6...2.8 GHz (average about 2.65 GHz)
running MBv8_8.05r3345_avx_linux64
throughput .................... 80,400 PPD (average from 200 valid tasks)
power consumption .......... 390 W (averaged by the watt meter during >1 day)
power efficiency .................. 2.4 points/kJ

GPU host: dual Pascal
2x GTX 1080Ti, watercooled, i7-7700K is only feeding the GPUs (i.e. no CPU tasks)
running setiathome_x41p_zi3v_x86_64-pc-linux-gnu_cuda90
throughput .................. 106,000 PPD (average from 200 valid tasks)
power consumption .......... 400 W (averaged by the watt meter during >1 day)
power efficiency .................. 3.1 points/kJ

I.e., these CPUs achieve 78 % of the efficiency of the GPU host, going by total host consumption at the wall.

Additional findings:

Coefficient of variation ( = standard deviation normalized to the average) of run times as well as of credits per task:
CPU host:
a bit more than 30 % with 100 tasks sample size,
a bit less than 30 % with 200 tasks sample size.​
four GPU hosts (all Pascal):
almost 40 % with 100 tasks sample size per host,
about 30 % with 200 tasks sample size per host.​

PPD have fallen compared to 2017 Wow!-Event:
The 80,400 PPD reported above are only 86 % of the 93,100 PPD which I measured last year on the same host with the same application.​

I also saw the decrease in PPD compared to the previous Wow!-Event briefly discussed in SETI.Germany's forum, though there were no hard numbers reported there.
 

Howdy

Senior member
Nov 12, 2017
572
480
136
Could some of the points deviation also be attributed to a crappy wing man? My graph along with Crash's and Lane's is all over the board, OTOH yours sees an upward trend with only slight variations. I have changed nothing besides turning on a fan here and there.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |