Posted 31 Dec 2020 by
Buro87 [Lombardia]
Almost certainly because the work generator cannot keep up with the demand.
As I recall, flat out it can generate a few thousand work units per day. At 24/7 your 1,024 threads would process around 8,000 work units.
If i remember correctly,what Valterc said while ago, work generator speed is around 1 gene every 14 minutes
A gene is composed by 294wu (*2 replicants)
So: 1440/14= 102 genes a day. 102*294*2= 59976wu daily
@ p3d-cluster: congratulations for your army. It's impressive to se theese dual EPYC monsters
Posted 24 Dec 2020 by
Buro87 [Lombardia]
Try to suspend others project, than ask work for TN-grid
If you want, the he following is the cc_config.xml i am using
<cc_config>
<options>
<allow_multiple_clients>1</allow_multiple_clients>
<max_file_xfers>50</max_file_xfers>
<max_file_xfers_per_project>30</max_file_xfers_per_project>
<report_results_immediately>1</report_results_immediately>
<ncpus>12</ncpus>
</options>
</cc_config>
set the ncpus value to your correct thread number: 32 (3950x), 16 (1700)
Others values are for multiple wu trasfer
By the way i write down also the app_config.xml file to put in the project folder (C:\ProgramData\BOINC\projects\gene.disi.unitn.it_test)
whit this you can specify the maximum number of concurrent wus. It's useful if you want run multiple projects
<app_config>
<project_max_concurrent>12</project_max_concurrent>
</app_config>
Posted 24 Dec 2020 by
Buro87 [Lombardia]
Tn-grid has limit of 8wu per core, and 512wu per pc, even with a Epyc (64c/128t)
With your 3950x you can reach 32x8= 256 running wu
Hello,
when do I get this 8 WU per Core?
I am running the project since some days now with more then 200 valid WUs but get only 1 WU per Core.
Its the same on 2 different PCs. The project was added completely new for both PCs.
Best Regards
Magiceye
I think you need to set your work buffer from your boinc manager preferences
with yor 1700 and 3950x setting 1day + 1day, is more than enough to reach 128wu and 256wu respectively
You need also to considerer the max concurrent wu limit.
If you have set, for example; a limit of 10 concurrent wu, trough an app_config.xml file, you will recieve 10*8=80 wu even with a 10+10days work buffer
Posted 15 Dec 2020 by
Buro87 [Lombardia]
Hi all
Are you ready? Tomorrow the challenge begins :) :)
Posted 9 Dec 2020 by
Buro87 [Lombardia]
When will be available for "normal" android devices?
Posted 7 Dec 2020 by
Buro87 [Lombardia]
i think you need to ask in the Boincstats forum.
i don't understand why you are in "Richard Dawkins fondation" (https://www.boincstats.com/stats/-5/user/detail/15233996966/overview)
but if you click on "project list -> TN-grid" you are part of "In science we trust" team (https://www.boincstats.com/stats/150/user/list/0/0#8)
which one is yor correct team?
Posted 2 Dec 2020 by
Buro87 [Lombardia]
Hi all
i'm here to report the fololowing challenge issued by Colombia team
https://www.boincstats.com/stats/challenge/team/chat/1084
TN-Grid December 2020 challenge
Start time: December 16th 2020 14:15 UTC
End time: December 23rd 2020 14:15 UTC
Posted 17 Nov 2020 by
Buro87 [Lombardia]
Don't worry. when challenge start, the wus will be validated :)
Posted 17 Nov 2020 by
Buro87 [Lombardia]
Hi all,
I have seen the following "Pirates anti covid weekend challenge" from November 20th to november 24th
https://www.boincstats.com/stats/challenge/team/chat/1083
Actually we are in "bunker phase" and the recent WUs' peak is probably due to this
Join the challenge and support TN-grid :)
Posted 14 Nov 2020 by
Buro87 [Lombardia]
If you have lots of pc is normal to have many pendig wu. It's usually take 3-4 days to have a stable pending WUs
And some users don't run their pc 24/7
I remember that, with my r5-1600 i have around 70 pendig WUs
By the way your pending credits will be assigned sooner or later. They will not be wasted :)
Posted 11 Nov 2020 by
Buro87 [Lombardia]
Hi Valterc
Any update regarding AMD HPC support?
Posted 11 Nov 2020 by
Buro87 [Lombardia]
Do you remember when I said is not good the "no work available" situacion?
Now, plenty of computers asked for work to some other projects and we moved from 60 genes / day to 44 genes / day.
I have seen than in my own machine.
Well, the server limits are obviously a big issue. However, the recent drop of performance here is mainly due to one single user (with an impressive number of hosts) that decided to stop/pause crunching (I don't see him active in other projects...)
On november 16th also start the annual WCG Birthday challenge, which is very popular
By the way in past days i have seen a top of 72 gene/day
Posted 4 Nov 2020 by
Buro87 [Lombardia]
Good news. How can we recognize theese WUs?
Posted 22 Oct 2020 by
Buro87 [Lombardia]
How much do they need? How many of us are there? I'm sure a lot of us would crowd fund it. It wouldn't be very expensive between 1000 people. SETI raised quite a bit of cash by asking folk.
They can't raise resources from crowdfunding because they are based in the Thrento University, so they need to follow the "official " way, which is more longer and buricratic
Posted 17 Oct 2020 by
Buro87 [Lombardia]
All server processes are running, yet no tasks available to send.
Out of work.
During the last day "running wu" are raised from 35k to 54k
Posted 27 Apr 2020 by
Buro87 [Lombardia]
Some 166977 completed :)
Posted 21 Apr 2020 by
Buro87 [Lombardia]
Hi Byron
welcome on TN-grid :)
Posted 18 Apr 2020 by
Buro87 [Lombardia]
The WCG also contains more competition. TN-Grid will not get a big chunk of the processing power just like that. It's not a free, unlimited resource. Even the WCG will set priorities and you might not like them.
Within boinc, it is your choice to do TN-Grid. Within WCG it is not.
Put TN-Grid on the official boinc project list, get a Twitter, Facebook, ... compaign going. Put posters up at local universities or wherever. And first upgrade the backend so the tn-grid doesn't end up like Rosetta (out of work and untested, badly failing WU in the wild) or F@H (haven't seen a WU in the last week's).
I agree. That' the point
TN-Grid, with their current servers, can't handle such a large number of users resulting from pubblication on the official boinc project list. As saod, they are planning to upgrade it. If all goes well, with new servers, with , maybe a Gpu app, and Android app, TN-Grid will scale up
For scientist a constant data flow is important.
Users want a full queue on their machine. Otherwise they get upset
Posted 17 Apr 2020 by
Buro87 [Lombardia]
@robertmiles
I don't understand your point.
TN-grid platform is is provided by Trnto University and National Research Council of Italy (CNR).
They are currently crunching on gene@home project
I think (and hope) they are planning to expand and add other projects to TN-grid platform in the future
Posted 10 Apr 2020 by
Buro87 [Lombardia]
Thank you very much L.Gilson :)