Trying to understand WU distribution.
Moderators: Site Moderators, FAHC Science Team
Multi GPU's
I have 2 GPU's but the software on seems to utilize one at a time. So out of 1 cpu and 2 gpus, it will only ever use 1.
Any ideas?
Thanks
Any ideas?
Thanks
-
- Posts: 1094
- Joined: Wed Nov 05, 2008 3:19 pm
- Location: Cambridge, UK
Re: Multi GPU's
As long as you have set up all three slots, they should all be used -- if the seervers can provide work fast enough. What does the log (in Advanced Control) say?
Re: Multi GPU's
Under slots it is currently showing three folding slots; 0 cpu, 1 gpu and 2 gpu
This is what I'm seeing in the log currently:-
#################
'18.218.241.186:80': No WUs available for this configuration
19:02:52:ERROR:WU02:FS00:Exception: Could not get an assignment
19:49:51:WARNING:WU02:FS00:Failed to get assignment from '65.254.110.245:8080': No WUs available for this configuration
19:49:51:WARNING:WU02:FS00:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration
19:49:51:ERROR:WU02:FS00:Exception: Could not get an assignment
19:52:43:WARNING:WU00:FS02:Failed to get assignment from '65.254.110.245:8080': No WUs available for this configuration
19:52:43:WARNING:WU00:FS02:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration
19:52:43:ERROR:WU00:FS02:Exception: Could not get an assignment
#################
Obviously you haven't included the full log.
I included warnings and errors, Presumably this means there isn't any work available at the moment?
Thanks
This is what I'm seeing in the log currently:-
#################
'18.218.241.186:80': No WUs available for this configuration
19:02:52:ERROR:WU02:FS00:Exception: Could not get an assignment
19:49:51:WARNING:WU02:FS00:Failed to get assignment from '65.254.110.245:8080': No WUs available for this configuration
19:49:51:WARNING:WU02:FS00:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration
19:49:51:ERROR:WU02:FS00:Exception: Could not get an assignment
19:52:43:WARNING:WU00:FS02:Failed to get assignment from '65.254.110.245:8080': No WUs available for this configuration
19:52:43:WARNING:WU00:FS02:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration
19:52:43:ERROR:WU00:FS02:Exception: Could not get an assignment
#################
Obviously you haven't included the full log.
I included warnings and errors, Presumably this means there isn't any work available at the moment?
Thanks
-
- Posts: 1164
- Joined: Wed Apr 01, 2009 9:22 pm
- Hardware configuration: Asus Z8NA D6C, 2 x5670@3.2 Ghz, , 12gb Ram, GTX 980ti, AX650 PSU, win 10 (daily use)
Asus Z87 WS, Xeon E3-1230L v3, 8gb ram, KFA GTX 1080, EVGA 750ti , AX760 PSU, Mint 18.2 OS
Not currently folding
Asus Z9PE- D8 WS, 2 E5-2665@2.3 Ghz, 16Gb 1.35v Ram, Ubuntu (Fold only)
Asus Z9PA, 2 Ivy 12 core, 16gb Ram, H folding appliance (fold only) - Location: Jersey, Channel islands
Re: Multi GPU's
You are partially correct, there is work available, the servers just cannot dish work out fast enough for everyone.
Re: Multi GPU's
viewtopic.php?f=18&t=32743&p=314523#p314523
Posting FAH's log:
How to provide enough info to get helpful support.
How to provide enough info to get helpful support.
Re: Multi GPU's
Yep I see. Fine I will leave it powered up and see
-
- Posts: 13
- Joined: Thu Mar 19, 2020 4:21 pm
Trying to understand WU distribution.
I have been folding on/off for a few years, but now with so many people helping, it is the first time I have tried it with not enough work for everyone.
In my main computer I got a Ryzen 3700x and a RTX2070, most of the day both cpu and gpu sits idle and isn't able to get any WU's to work on.
My server is based on an old 4790k with a Quadro P400 card in, it gets WU's almost instantly.
Is there some kind of balancing going on here depending on hardware, or is it just pure luck it is able to get them so fast? (both cpu and gpu in my server gets WU's fast)
In my main computer I got a Ryzen 3700x and a RTX2070, most of the day both cpu and gpu sits idle and isn't able to get any WU's to work on.
My server is based on an old 4790k with a Quadro P400 card in, it gets WU's almost instantly.
Is there some kind of balancing going on here depending on hardware, or is it just pure luck it is able to get them so fast? (both cpu and gpu in my server gets WU's fast)
Re: Trying to understand WU distribution.
it is pure luck combined with actually getting WU's means it resets the time between checking of available WU's. Every time an attempt is made to see if an WU is available the time will go up before it checks again, from 1 minute to 5 minutes etc. You might want to see the "next attempt" timer on the advanced control client on your main computer, the way to reset it is by pausing and resuming but do not overuse it as it will put unnecessary stress on the WS servers
-
- Site Moderator
- Posts: 2850
- Joined: Mon Jul 18, 2011 4:44 am
- Hardware configuration: OS: Windows 10, Kubuntu 19.04
CPU: i7-6700k
GPU: GTX 970, GTX 1080 TI
RAM: 24 GB DDR4 - Location: Western Washington
Re: Trying to understand WU distribution.
It some hardware but mostly luck. There are so many new volunteers joining F@h over the past several days that the servers and queue of workunits has been overwhelmed. The research teams are rapidly trying to set up new servers and add new projects into the queue. In the meantime, you should be able to pick up workunits when they are available.
F@h is now the top computing platform on the planet and nothing unites people like a dedicated fight against a common enemy. This virus affects all of us. Lets end it together.
-
- Posts: 13
- Joined: Thu Mar 19, 2020 4:21 pm
Re: Trying to understand WU distribution.
The pausing/unpausing trick is something I am already doing when the retry timer gets over an hour. I was just wondering because my older computer seems to get plenty of work.Harmin wrote:it is pure luck combined with actually getting WU's means it resets the time between checking of available WU's. Every time an attempt is made to see if an WU is available the time will go up before it checks again, from 1 minute to 5 minutes etc. You might want to see the "next attempt" timer on the advanced control client on your main computer, the way to reset it is by pausing and resuming but do not overuse it as it will put unnecessary stress on the WS servers
Re: Trying to understand WU distribution.
as far as I know it is just pure luck then
Re: Trying to understand WU distribution.
I'm discovering that there are situations where luck works and (apparently) where luck doesn't work as it "should" I'm trying to figure out why a number of my slots which have been idle for way too long are now folding and others which have also been idle still don't have work. The fact is, "FAH-Central" brought considerable additional server resources on-line within the past day or so and filled them with new WUs, so the problems are being solved. ... (Better, but not yet "solved"). Nevertheless, in this time of high system stress, some software deficiencies that we've always had may get exposed and if I can diagnose them, maybe there's room for improvment for a couple of things that simply don't show up under less stressing conditions.
I'm not sure I'd use the word "pure" in front of the word "luck" but it really is mostly luck.
I'm not sure I'd use the word "pure" in front of the word "luck" but it really is mostly luck.
Posting FAH's log:
How to provide enough info to get helpful support.
How to provide enough info to get helpful support.
Re: Trying to understand WU distribution.
I put as far as I know it infront of that since I have no idea what goes around on the back end but that sounds plausible
Re: Trying to understand WU distribution.
As i interpret these stats...
The servers are handing out 70K WUs per hour. Which is a bump from 55K earlier this week, ( no clue what number was prior to coronatimes )
So there is a lot of work being assigned, but contrary to what the error message in the client is saying ( WARNING:WU00:FS01:Failed to get assignment from 'xx.xxx.xxx.xxx:80': No WUs available for this configuration )
There are WUs available, it's just that there are more clients asking for them than the servers currently have capacity to talk to.
That's just my guess, i have no real insight into the matters and could be wrong.
The servers are handing out 70K WUs per hour. Which is a bump from 55K earlier this week, ( no clue what number was prior to coronatimes )
So there is a lot of work being assigned, but contrary to what the error message in the client is saying ( WARNING:WU00:FS01:Failed to get assignment from 'xx.xxx.xxx.xxx:80': No WUs available for this configuration )
There are WUs available, it's just that there are more clients asking for them than the servers currently have capacity to talk to.
That's just my guess, i have no real insight into the matters and could be wrong.
Re: Multi GPU's
According to https://apps.foldingathome.org/serverstats, FAH has 9 active Work Servers. Some have projects for GPUs, some have projects for CPUs, and a few have both. Right now I see that four of them are delivering 10,802.21/hr which must be some kind of capacity limitation. Four more are delivering 3,602.21/hr ... presumably an older/slower server with more limitations. One more which isn't doing much at the moment.Rickit wrote:Under slots it is currently showing three folding slots; 0 cpu, 1 gpu and 2 gpu
This is what I'm seeing in the log currently:-...
[Note: Available jobs and assign rates are estimated.]
Those estimated transaction rates (xxxxxx/hr) probably include both events where a WU was downloaded and events where no WU was found that matched your client's requirements. I'm not sure where the upload transactions appear on that report.
Your client went to "assign1" ('65.254.110.245) to look for a Work Server that had WUs and then to '18.218.241' (which is new and isn't even listed on that app, but I'm going to tuess it's another Assign Server). You were told "Exception: Could not get an assignment" That is the message you get when all the Work Servers don't have GPU WUs (if you're looking for a CPU WU) or don't have CPU WUs if you're looking for a GPU WU. The same message pops up if all the servers are empty or all of them are down, but I'm going to conclude that it also pops up if all of them are unable to service your request due to the fact that they are operating at their maximum capacity.
NOTE: Merged two topics discussing (more or less) the same subject. (Check the topic name on each post to review things that weren't here a minute ago)
Posting FAH's log:
How to provide enough info to get helpful support.
How to provide enough info to get helpful support.