Advanced search

Message boards : Number crunching : No work for me to crunch..

Author Message
Mads Nissen
Send message
Joined: 23 Jun 11
Posts: 1
Credit: 82,578,269
RAC: 115,584
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwat
Message 53448 - Posted: 19 Jan 2020 | 19:46:00 UTC

Hi all, for a long time now, i have not seen/got any work for my pc to work with, is it only me or....

Nick Name
Send message
Joined: 3 Sep 13
Posts: 51
Credit: 1,371,556,456
RAC: 251,958
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 53454 - Posted: 21 Jan 2020 | 6:03:19 UTC - in response to Message 53448.

It's not just you. Right now we have two "problems":

1) There's not a lot of work.
2) There are a lot of crunchers trying to get work.

I still would have expected you to get some work in the last few days though. It may be that you have too much work from other projects when there is work here. Check your other project resource shares, you might need to lower those or increase GPUGrid's share, or both if you want to get more work here.
____________
Team USA forum | Team USA page
Join us and #crunchforcures. We are now also folding:join team ID 236370!

Profile microchip
Avatar
Send message
Joined: 4 Sep 11
Posts: 110
Credit: 303,630,118
RAC: 45,769
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 53455 - Posted: 21 Jan 2020 | 16:21:28 UTC

I was able to crunch one WU yesterday... and then nothing more
____________

Team Belgium

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 198
Credit: 1,456,827,348
RAC: 912,695
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 53458 - Posted: 23 Jan 2020 | 13:33:29 UTC

Sometimes, a scarcity of tasks like current one antecedes a “change of cycle” in GPUGrid...

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 508
Credit: 525,703,472
RAC: 1,674,796
Level
Lys
Scientific publications
wat
Message 53461 - Posted: 25 Jan 2020 | 9:02:45 UTC - in response to Message 53458.

Sometimes, a scarcity of tasks like current one antecedes a “change of cycle” in GPUGrid...

Seems you are prescient. The work is coming in like gangbusters.

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 198
Credit: 1,456,827,348
RAC: 912,695
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 53464 - Posted: 25 Jan 2020 | 11:45:29 UTC - in response to Message 53461.

It's not been a premonition, it was a Déjà vu :-)

marsinph
Send message
Joined: 11 Feb 18
Posts: 30
Credit: 289,515,345
RAC: 702,487
Level
Asn
Scientific publications
wat
Message 53468 - Posted: 25 Jan 2020 | 16:11:13 UTC

Server status shows increasing "ready to send WU".
On 25jan 16:00GMT above four thousands. Nice
By manual update the logfile report no rask !!!
My cache is full empty (only one Yafu WU running.
Ressource share at 200%
Host : http://www.gpugrid.net/show_host_detail.php?hostid=466817


25/01/2020 17:05:55 | GPUGRID | update requested by user
25/01/2020 17:05:59 | GPUGRID | Sending scheduler request: Requested by user.
25/01/2020 17:05:59 | GPUGRID | Requesting new tasks for CPU and NVIDIA GPU
25/01/2020 17:06:01 | GPUGRID | Scheduler request completed: got 0 new tasks
25/01/2020 17:06:01 | GPUGRID | No tasks sent
25/01/2020 17:06:01 | GPUGRID | No tasks are available for Short runs (2-3 hours on fastest card)
25/01/2020 17:06:01 | GPUGRID | No tasks are available for Long runs (8-12 hours on fastest card)
25/01/2020 17:06:01 | GPUGRID | No tasks are available for New version of ACEMD
25/01/2020 17:06:36 | GPUGRID | Sending scheduler request: To fetch work.
25/01/2020 17:06:36 | GPUGRID | Requesting new tasks for CPU and NVIDIA GPU
25/01/2020 17:06:37 | GPUGRID | Scheduler request completed: got 0 new tasks
25/01/2020 17:06:37 | GPUGRID | No tasks sent
25/01/2020 17:06:37 | GPUGRID | No tasks are available for Short runs (2-3 hours on fastest card)
25/01/2020 17:06:37 | GPUGRID | No tasks are available for Long runs (8-12 hours on fastest card)
25/01/2020 17:06:37 | GPUGRID | No tasks are available for New version of ACEMD

Aurum
Avatar
Send message
Joined: 12 Jul 17
Posts: 252
Credit: 9,791,563,847
RAC: 3,936,285
Level
Tyr
Scientific publications
wat
Message 53476 - Posted: 25 Jan 2020 | 20:09:14 UTC - in response to Message 53468.

How do you get full of empty???

marsinph
Send message
Joined: 11 Feb 18
Posts: 30
Credit: 289,515,345
RAC: 702,487
Level
Asn
Scientific publications
wat
Message 53477 - Posted: 25 Jan 2020 | 20:47:51 UTC - in response to Message 53476.

How do you get full of empty???

Mister; I not ask remarks about my english.

Full of empty .
It is language :
Full, fully = totally = whole ...
Empty : nothing

In french : totalement vide
Google translation : totally empty (perhaps Google is wrong !?)
See https://translate.google.be/?hl=fr#view=home&op=translate&sl=fr&tl=en&text=totalement%20vide

I suggest you try to write french like me in english.
You perfect know who I am . I perfect know who you are.
I not forget your expulsion, latest year, from GridCoin ...
Go on SETIBZH , take the first pace everywhere. You should show...
I not need language lessons.
I search answer, technical answer ! Not "languages courses"
Not Me and my team are not a danger with all your more than 600 cores and your about 70 GPU GTX 1080Ti/2080-2080Ti
Thank you.

By the way, my question is about server status and report.

Aurum
Avatar
Send message
Joined: 12 Jul 17
Posts: 252
Credit: 9,791,563,847
RAC: 3,936,285
Level
Tyr
Scientific publications
wat
Message 53494 - Posted: 26 Jan 2020 | 5:54:28 UTC - in response to Message 53477.

You so funny.
BTW, I like peanut butter on my Belgian waffles, it fills the empty.

marsinph
Send message
Joined: 11 Feb 18
Posts: 30
Credit: 289,515,345
RAC: 702,487
Level
Asn
Scientific publications
wat
Message 53507 - Posted: 26 Jan 2020 | 17:48:16 UTC - in response to Message 53494.

You so funny.
BTW, I like peanut butter on my Belgian waffles, it fills the empty.




Ah ah ah ah !!!

BTW in Belgium, means : VAT (additional taxes)
Strange you speak about "waffles". Very strange...
It remember me something in 2018...
Here, on our waffles, we prefer ice cream, marmelade, fruits, something not full fat like peanuts butter. We try to take care about our BMI (body mass index)

So after those considereations, about food, my hosts needs food from GPUGrid...
Or perhaps all available WU are in stock on your hundreds of GPU !?
It would explain your CR (the sixth, all around the world, all team together.


Miklos M.
Send message
Joined: 16 Jun 12
Posts: 17
Credit: 132,739,384
RAC: 1,364
Level
Cys
Scientific publications
watwat
Message 53621 - Posted: 7 Feb 2020 | 21:11:58 UTC

Where are the wu's?

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 508
Credit: 525,703,472
RAC: 1,674,796
Level
Lys
Scientific publications
wat
Message 53622 - Posted: 8 Feb 2020 | 1:33:27 UTC - in response to Message 53621.

Where are the wu's?

Sure aren't many. I get about one a day for the past week.

Miklos M.
Send message
Joined: 16 Jun 12
Posts: 17
Credit: 132,739,384
RAC: 1,364
Level
Cys
Scientific publications
watwat
Message 53623 - Posted: 8 Feb 2020 | 2:36:14 UTC - in response to Message 53622.

I guess I am paying the piper for those 151 wu's i got a while back in 1 day.

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 198
Credit: 1,456,827,348
RAC: 912,695
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 53624 - Posted: 8 Feb 2020 | 11:19:04 UTC

Where are the wu's?

Sure aren't many. I get about one a day for the past week.

I guess I am paying the piper for those 151 wu's i got a while back in 1 day.

I'm assuming the current pattern for this GPUGrid's stage: cycling 1-2 days of plentyness, then 1-2 weeks of scarcity.
It seems to be intrinsic to the kind of study we are supporting now.
As in previous times, I'll be proud to contribute to whatever scientific discoveries they may lead to.

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 198
Credit: 1,456,827,348
RAC: 912,695
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 53625 - Posted: 8 Feb 2020 | 15:44:46 UTC - in response to Message 53624.

I'm assuming the current pattern for this GPUGrid's stage: cycling 1-2 days of plentyness, then 1-2 weeks of scarcity.

In fact, plentyness periods would be longer and scarcity ones shorter if every computer downloaded at every moment no more than the maximum number of WUs it is able to process in one or two working days...
It is something not so easy to control on the Project's side.
However, it is possible for we crunchers to do a lot from our side by putting a little effort to autocontrol.
But probably it is not so easy neither...

Werinbert
Send message
Joined: 12 May 13
Posts: 2
Credit: 32,668,431
RAC: 482,887
Level
Val
Scientific publications
wat
Message 53626 - Posted: 8 Feb 2020 | 18:02:13 UTC

Economics...Supply and demand.

People have a perception of a lack of supply so they hoard which consequently cause a lack of supply. Self-fulfilling prophecy. Unfortunately it is hard to overcome the human nature of greed.

Erich56
Send message
Joined: 1 Jan 15
Posts: 697
Credit: 3,294,399,981
RAC: 381,374
Level
Arg
Scientific publications
watwatwatwatwatwat
Message 53627 - Posted: 8 Feb 2020 | 19:17:53 UTC - in response to Message 53624.

I'm assuming the current pattern for this GPUGrid's stage: cycling 1-2 days of plentyness, then 1-2 weeks of scarcity.

not always was it like that.
Before, there were many weeks of plentyness, and only from time to time some short periods of scarcity came up.
With the introduction of the new ACEMD, this pattern was changed, for whatever reason.

As mentioned recently, this unfortunately results in a lot more "thermal cycles" for the GPUs which reduces their lifetime. Too bad.

Nick Name
Send message
Joined: 3 Sep 13
Posts: 51
Credit: 1,371,556,456
RAC: 251,958
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 53628 - Posted: 9 Feb 2020 | 0:40:53 UTC

There are a number of things that are affecting work availability.

    1. There are fewer projects/researchers sending work.
    2. Tasks are smaller than in the past.
    3. Smaller tasks = faster run times, and this is worsened as Turing cards chew through these tasks so quickly. This means demand is greatly increased compared to the past.



All these combined mean it's harder to get work. We used to have at least 2-3 students/researchers at a time sending work. Hopefully that will happen again soon.

____________
Team USA forum | Team USA page
Join us and #crunchforcures. We are now also folding:join team ID 236370!

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2185
Credit: 15,824,047,857
RAC: 697,719
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 53631 - Posted: 9 Feb 2020 | 19:16:12 UTC - in response to Message 53626.

Economics...Supply and demand.

People have a perception of a lack of supply so they hoard which consequently cause a lack of supply. Self-fulfilling prophecy. Unfortunately it is hard to overcome the human nature of greed.
True for the economy, but we're volunteers.
Greedy volunteers... Volunteers demanding work... both of that sounds weird.
It comes from that the volunteers' perception of their relation with the project is wrong.
It's actually the way around:
There's a very large supply of free computing power from the volunteer's part, but there's a low demand for it from GPUGrid's part.
My advice of the day:
If the mountain will not go to Mahomet, Mahomet will go to the mountain.
e.g. set up (attach to) a backup GPU project with 0 (zero) resource share.
This way the BOINC manager will always download only a single workunit (from the backup project), regardless of the cache settings.

Post to thread

Message boards : Number crunching : No work for me to crunch..