Advanced search

Message boards : News : Doubled number of hosts

Author Message
Toni
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 9 Dec 08
Posts: 947
Credit: 4,353,973
RAC: 1,134
Level
Ala
Scientific publications
watwatwatwat
Message 53999 - Posted: 23 Mar 2020 | 12:43:11 UTC
Last modified: 23 Mar 2020 | 12:44:23 UTC

If I am not missing anything, GPUGRID volunteers doubled the number of active hosts in the last month. It may have been either due to the closing of seti@home, or a consequence of folding at home’s covid effort.



(Credits: Boincstats)

Felix_M_
Send message
Joined: 29 Jan 16
Posts: 9
Credit: 14,546,498
RAC: 13,193
Level
Pro
Scientific publications
wat
Message 54000 - Posted: 23 Mar 2020 | 13:03:00 UTC

we will see how many of them will stay permanently, but great news

kain
Send message
Joined: 3 Sep 14
Posts: 150
Credit: 562,363,527
RAC: 163,666
Level
Lys
Scientific publications
watwatwatwatwatwatwatwat
Message 54001 - Posted: 23 Mar 2020 | 14:42:55 UTC

Well, now we have plenty of work units. It's very important.

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 500
Credit: 474,948,762
RAC: 1,459,694
Level
Gln
Scientific publications
wat
Message 54002 - Posted: 23 Mar 2020 | 15:41:15 UTC

I expected such. Going to be more hosts coming over from Seti@home with the project finishing up. Soon as the project stops sending out work, all my gpus will be doing GPUGrid full time with higher resource shares.

Jim1348
Send message
Joined: 28 Jul 12
Posts: 731
Credit: 1,474,393,613
RAC: 43,196
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 54005 - Posted: 23 Mar 2020 | 18:28:58 UTC - in response to Message 54002.

Going to be more hosts coming over from Seti@home with the project finishing up.

Dave Anderson is a genius. He stopped SETI at just the right time.

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 500
Credit: 474,948,762
RAC: 1,459,694
Level
Gln
Scientific publications
wat
Message 54006 - Posted: 23 Mar 2020 | 21:18:01 UTC - in response to Message 54005.

Going to be more hosts coming over from Seti@home with the project finishing up.

Dave Anderson is a genius. He stopped SETI at just the right time.

I have another rude description for DA.

Zalster
Avatar
Send message
Joined: 26 Feb 14
Posts: 208
Credit: 4,490,828,031
RAC: 194,498
Level
Arg
Scientific publications
watwatwatwat
Message 54007 - Posted: 23 Mar 2020 | 21:24:01 UTC - in response to Message 54006.
Last modified: 23 Mar 2020 | 21:24:31 UTC

+1

mmonnin
Send message
Joined: 2 Jul 16
Posts: 284
Credit: 817,372,805
RAC: 1,268,382
Level
Glu
Scientific publications
wat
Message 54008 - Posted: 23 Mar 2020 | 23:24:50 UTC
Last modified: 23 Mar 2020 | 23:47:24 UTC

I used to have a GTX 1080 dedicated to FAH but there has been limited work there so I moved it to GPUGrid.

Edit: I have pointed some team members here for GPUs and Rosetta for CPUs.

Rosetta has taken off as well
https://stats3.free-dc.org/stats.php?page=proj&proj=rah

[CSF] Thomas H.V. Dupont
Send message
Joined: 20 Jul 14
Posts: 727
Credit: 96,837,728
RAC: 19,130
Level
Thr
Scientific publications
watwatwatwatwatwat
Message 54009 - Posted: 24 Mar 2020 | 7:05:42 UTC - in response to Message 54001.

Well, now we have plenty of work units. It's very important.

It's certainly linked to that indeed.
We stayed for a very long time in shortage of WU's which had to effect inactivating a lot of hosts statistically speaking.
____________
[CSF] Thomas H.V. Dupont
Founder of the team CRUNCHERS SANS FRONTIERES
www.crunchersansfrontieres.org

[AF>Le_Pommier] Jerome...
Send message
Joined: 12 May 17
Posts: 2
Credit: 10,293,450
RAC: 0
Level
Pro
Scientific publications
wat
Message 54011 - Posted: 24 Mar 2020 | 10:02:27 UTC

Without DA none of us would be here. We would fold, maybe :)

The problem with GPUGrid is that it is often short of tasks to process, I hope that the current trend is permanent.

root
Send message
Joined: 23 Mar 20
Posts: 1
Credit: 442,913
RAC: 48
Level

Scientific publications
wat
Message 54012 - Posted: 24 Mar 2020 | 10:40:00 UTC - in response to Message 53999.

most for sure the corvid19, that's why i'm here.
also the foh gpu client i was running was not really receiving many work units... so i switched away to here.

"cant we all just get along" :)

Mr. Kevvy
Avatar
Send message
Joined: 1 Jun 16
Posts: 12
Credit: 559,663,678
RAC: 8,205,505
Level
Lys
Scientific publications
wat
Message 54014 - Posted: 24 Mar 2020 | 12:10:21 UTC - in response to Message 53999.
Last modified: 24 Mar 2020 | 12:20:22 UTC

For me ... and a couple of my teammates from SETI@Home -- Hey guys! :^) ... it's the closure. I picked this project because it's both a worthy and certainly important cause especially these days, and because it has a CUDA 10 Linux client. No point in wasting the GPUs' potential.

You'll doubtless see quite a few more "resetilers" migrating here with our RTX farms in the near future.

Atri
Send message
Joined: 15 Mar 20
Posts: 4
Credit: 752,098
RAC: 81
Level
Gly
Scientific publications
wat
Message 54015 - Posted: 24 Mar 2020 | 12:11:19 UTC - in response to Message 53999.

ciao a tutti vi scrivo dall'italia. Chiedevo una informazione: Gpu-Grid sta eseguendo progetti x il Coronavirus Covid-19?

Atri
Send message
Joined: 15 Mar 20
Posts: 4
Credit: 752,098
RAC: 81
Level
Gly
Scientific publications
wat
Message 54016 - Posted: 24 Mar 2020 | 12:16:28 UTC - in response to Message 53999.

If I am not missing anything, GPUGRID volunteers doubled the number of active hosts in the last month. It may have been either due to the closing of seti@home, or a consequence of folding at home’s covid effort.



(Credits: Boincstats)


Gpu- Grid esegue progetti x il coronavirus Covid19?

victorA
Send message
Joined: 30 May 14
Posts: 3
Credit: 23,372,178
RAC: 38,289
Level
Pro
Scientific publications
wat
Message 54017 - Posted: 24 Mar 2020 | 12:21:31 UTC - in response to Message 53999.

For me its working from home. For sure im giving you more cpu!

[VENETO] boboviz
Send message
Joined: 10 Sep 10
Posts: 142
Credit: 388,132
RAC: 0
Level

Scientific publications
wat
Message 54020 - Posted: 24 Mar 2020 | 15:11:12 UTC - in response to Message 54006.

Dave Anderson is a genius. He stopped SETI at just the right time.

I have another rude description for DA.


Why?

[VENETO] boboviz
Send message
Joined: 10 Sep 10
Posts: 142
Credit: 388,132
RAC: 0
Level

Scientific publications
wat
Message 54021 - Posted: 24 Mar 2020 | 15:17:26 UTC - in response to Message 54016.

Gpu- Grid esegue progetti x il coronavirus Covid19?

Per ora no, ma ci stanno lavorando.

ctnfoster
Send message
Joined: 8 Jul 08
Posts: 1
Credit: 7,194,248
RAC: 112,392
Level
Ser
Scientific publications
watwatwat
Message 54022 - Posted: 24 Mar 2020 | 15:17:43 UTC - in response to Message 53999.

As someone who recently rejoined, GPUGRID is the only project issuing sufficient work units for GPUs right now. Everyone else is CPU bound or issuing GPU work units not compatible with my hardware.

Thorstel
Send message
Joined: 16 Mar 20
Posts: 1
Credit: 19,208,525
RAC: 6,029
Level
Pro
Scientific publications
wat
Message 54023 - Posted: 24 Mar 2020 | 15:53:45 UTC - in response to Message 53999.

I am here because of all the folding@home news on gaming/hardware related websites. I've tried it but it often doesn't have any work for me since everybody is doing it now so I've looked for alternatives.

I really liked the idea of donating my processing power and even thought about it many times before but never came to research the idea for some reason.

I am typically running high-end gaming setup with dual GPU and high-end HEDT CPU that even if I play a lot still sits unused most of the day, and since there isn't really any reliable way for such system to make me any reasonable amount of money, I may just as well donate the processing power to charity.

This is pretty genius way of reaching so many people who otherwise do not donate money for charity for multiple reasons, donating something that you already have in unlimited amount, I am surprised this isn't more common and that I've heard about it only now.

Even from hardware enthusiast standpoint this is fun because you have to optimize your system also for long-term full load operation and energy efficiency rather than only squeezing every last MHz like for gaming and overclocking, and you want to be able to switch between these two settings on the fly depending on what you are about to do, so that has many interesting challenges in it's own.

GabrielB
Send message
Joined: 16 Mar 20
Posts: 3
Credit: 10,420,273
RAC: 173,328
Level
Pro
Scientific publications
wat
Message 54026 - Posted: 24 Mar 2020 | 17:37:46 UTC

I'm here because F@H caused me to look at my BOINC processes and I saw I had underutilized GPU resources. It doesn't hurt that SETI is done for now.

WorldComposting
Send message
Joined: 16 Mar 20
Posts: 1
Credit: 13,596,506
RAC: 144,101
Level
Pro
Scientific publications
wat
Message 54027 - Posted: 24 Mar 2020 | 19:02:16 UTC

I recently joined after doing work on World Community Grid for year and realized my GPU was not being used. I installed F@H and I wasn't getting any work units.

I then went into BOINC manager on my PC and realized that GPUGRID was meant for GPU usage.

I really think F@H needs to work in BOINC so that way if units are not being sent out you can do other work.

lakotamm
Send message
Joined: 6 Jul 19
Posts: 1
Credit: 841,909
RAC: 151
Level
Gly
Scientific publications
wat
Message 54029 - Posted: 24 Mar 2020 | 20:47:01 UTC

I discovered BOINC in June and joined WCG in July. Have been running it since then. I tried to run GPU GRID and F@H, but it seemed like the GPU was overheating my tiny T490 way too much.
Due to Covid 19, I revised the situation and found out that I can get my GPU to run reliably with GPU GRID while throttled to ca 60%. And so I am here, running the work units with my tiny GPU. F@H does not have enough work units and it seems like it often fails for some reason. GPU GRID WUs feel much more "stable".

I do not mind staying here, but it seems like it is almost impossible to figure out what is the current research used for. I do not mind working on other causes than COVID 19, but it has to be something useful, right?

MoeDrippins
Send message
Joined: 23 Mar 20
Posts: 1
Credit: 5,808,801
RAC: 60,814
Level
Ser
Scientific publications
wat
Message 54030 - Posted: 24 Mar 2020 | 21:06:57 UTC

> closing of seti@home, or a consequence of folding at home’s covid effort.

You have guessed the 2 reasons I joined.

Pop Piasa
Avatar
Send message
Joined: 8 Aug 19
Posts: 91
Credit: 104,518,948
RAC: 700,539
Level
Cys
Scientific publications
wat
Message 54033 - Posted: 24 Mar 2020 | 22:20:15 UTC

I recently doubled up on GPUs in my 2 puny hosts, does that count?

Pop Piasa
Avatar
Send message
Joined: 8 Aug 19
Posts: 91
Credit: 104,518,948
RAC: 700,539
Level
Cys
Scientific publications
wat
Message 54036 - Posted: 24 Mar 2020 | 23:47:09 UTC - in response to Message 54033.
Last modified: 25 Mar 2020 | 0:16:40 UTC

[Deleted]

Pop Piasa
Avatar
Send message
Joined: 8 Aug 19
Posts: 91
Credit: 104,518,948
RAC: 700,539
Level
Cys
Scientific publications
wat
Message 54037 - Posted: 24 Mar 2020 | 23:48:12 UTC - in response to Message 54033.
Last modified: 25 Mar 2020 | 0:15:54 UTC

{Deleted}

Pop Piasa
Avatar
Send message
Joined: 8 Aug 19
Posts: 91
Credit: 104,518,948
RAC: 700,539
Level
Cys
Scientific publications
wat
Message 54038 - Posted: 24 Mar 2020 | 23:52:05 UTC - in response to Message 54033.

I recently doubled up on GPUs in my 2 puny hosts, does that count?

Sarcastic humor aside, congratulations to Toni and the staff on the uptick in available hosts. It shows that this is an undertaking very worthy of the support of the global crunching community.

Distributed computing is a way everyone can participate in prevention and cure, no matter what the current biological villain might be. Yet, medicine is only one of the many applications of distributed computing. Long live DC and its spirit of volunteerism!

Pop Piasa
Avatar
Send message
Joined: 8 Aug 19
Posts: 91
Credit: 104,518,948
RAC: 700,539
Level
Cys
Scientific publications
wat
Message 54040 - Posted: 25 Mar 2020 | 0:01:48 UTC

Appologies for the multiple posts. I am having extreme delays in posting.

du-china
Send message
Joined: 16 Mar 20
Posts: 1
Credit: 418,106
RAC: 29
Level

Scientific publications
wat
Message 54041 - Posted: 25 Mar 2020 | 1:10:21 UTC

可喜可贺

NyeLabs
Send message
Joined: 31 Dec 15
Posts: 1
Credit: 44,020,836
RAC: 174,198
Level
Val
Scientific publications
wat
Message 54043 - Posted: 25 Mar 2020 | 4:25:42 UTC

I have had a few PCs on gpugrid for some time. With the ending of the SETI workload I have started adding gpugrid to all my GPU systems. The point about the Covid-19 virus work is added incentive. Some older or lower end cards that would do SETI work do not get gpugrid work allocated to them. So some upgrades will be needed in the future, probably post recession...

rod4x4
Send message
Joined: 4 Aug 14
Posts: 158
Credit: 1,825,201,071
RAC: 977,598
Level
His
Scientific publications
watwatwatwatwatwatwat
Message 54044 - Posted: 25 Mar 2020 | 4:28:39 UTC - in response to Message 54043.

I have had a few PCs on gpugrid for some time. With the ending of the SETI workload I have started adding gpugrid to all my GPU systems. The point about the Covid-19 virus work is added incentive. Some older or lower end cards that would do SETI work do not get gpugrid work allocated to them. So some upgrades will be needed in the future, probably post recession...


Updated FAQ for GPU requirements can be found at this link:
http://www.gpugrid.net/forum_thread.php?id=5002

sombraguerrero
Send message
Joined: 10 Dec 16
Posts: 1
Credit: 33,961,569
RAC: 475,943
Level
Val
Scientific publications
wat
Message 54045 - Posted: 25 Mar 2020 | 5:19:55 UTC

I've actually been joined to this project for some time, but yes, I am primarily a FAH participant. Since their influx has actually ended up pushing a lot of mid-upper range rigs out of the pool, I came back here. I have to admit, part of the reason I didn't stay initially was because the configs are a little cumbersome, until you take the time to get accustomed to synchronizing between web and local.

XeroOne
Send message
Joined: 17 Mar 20
Posts: 1
Credit: 2,446,855
RAC: 301
Level
Ala
Scientific publications
wat
Message 54046 - Posted: 25 Mar 2020 | 5:20:48 UTC
Last modified: 25 Mar 2020 | 5:31:11 UTC

I used BOINC many years back under a different name, but for some reason I never installed it on this Gaming Rig.

After SpiceWorks announced the Covid-19 F@H effort I signed up to F@H, but due to lack of units I came back home to BOINC and GPUGRD/Rosetta/WCG.

So here I am.

Chris Rampson
Send message
Joined: 8 Apr 09
Posts: 3
Credit: 211,339,674
RAC: 117,001
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54053 - Posted: 25 Mar 2020 | 11:50:13 UTC
Last modified: 25 Mar 2020 | 11:51:25 UTC

I receive maybe 1 or 2 units per MONTH. What are these new workunits you speak of?

[AF>Le_Pommier] Jerome...
Send message
Joined: 12 May 17
Posts: 2
Credit: 10,293,450
RAC: 0
Level
Pro
Scientific publications
wat
Message 54055 - Posted: 25 Mar 2020 | 13:58:14 UTC
Last modified: 25 Mar 2020 | 13:59:46 UTC

Reacting to some comments above, in case neither folding nor GPUGrid have work for GPU at the moment there are several other boinc projects with GPU work to be done, but I think none in the biomed area, but you have

Mathematics :

Moo! Wrapper
PrimeGrid
Collatz
Amicable
NumberFields (but I think not more efficient than their CPU app, but still)


Astronomy :

MilkyWay
Einstein
Asteroid

Be careful with those having OpenCL apps, they often require one full CPU core to be able to feed the GPU app with work (means that other CPU boinc app running at the same time will have one core less for them, but still).

Gresse Olivier
Send message
Joined: 15 Oct 15
Posts: 3
Credit: 24,298,046
RAC: 13,264
Level
Pro
Scientific publications
watwatwatwat
Message 54059 - Posted: 25 Mar 2020 | 15:44:08 UTC

Team,

My laptops, were crunching for S@H with their small GPU (no power loss) but they are not getting GPUGrid tasks...

Any plan to catch up and grab this computing power as well for the Grid ?

Thanks
Oli

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 500
Credit: 474,948,762
RAC: 1,459,694
Level
Gln
Scientific publications
wat
Message 54060 - Posted: 25 Mar 2020 | 15:52:12 UTC - in response to Message 54059.

Team,

My laptops, were crunching for S@H with their small GPU (no power loss) but they are not getting GPUGrid tasks...

Any plan to catch up and grab this computing power as well for the Grid ?

Thanks
Oli

Read the FAQ.
https://www.gpugrid.net/forum_thread.php?id=5002

Pay attention to this part.

I get errors, and the output includes "CUDA_ERROR_UNKNOWN (999)"

You may have a notebook (or older) graphic card. They are not really suited to sustain hard-core computations (thanks anyway).


Laptop iGPUs don't have the power or enough memory to crunch the new acemd3 application.

Wilgard
Send message
Joined: 4 Mar 20
Posts: 11
Credit: 1,774,754
RAC: 12,255
Level
Ala
Scientific publications
wat
Message 54065 - Posted: 25 Mar 2020 | 16:51:54 UTC - in response to Message 53999.

I just join the project and it is to fight against the COVID-19 :)

Aurum
Avatar
Send message
Joined: 12 Jul 17
Posts: 247
Credit: 9,673,488,208
RAC: 7,293,473
Level
Tyr
Scientific publications
wat
Message 54067 - Posted: 25 Mar 2020 | 22:52:31 UTC - in response to Message 54027.
Last modified: 25 Mar 2020 | 22:52:50 UTC

I really think F@H needs to work in BOINC so that way if units are not being sent out you can do other work.

Those Stanford guys are too arrogant to use BOINC. I've been telling them that for years. It's hopeless.

Gresse Olivier
Send message
Joined: 15 Oct 15
Posts: 3
Credit: 24,298,046
RAC: 13,264
Level
Pro
Scientific publications
watwatwatwat
Message 54068 - Posted: 25 Mar 2020 | 22:53:35 UTC - in response to Message 54060.
Last modified: 25 Mar 2020 | 22:58:12 UTC

Thx for the reply and pointing me to the right article.

Seems the laptops will stay very cool... :(
Unless someone has a trick ?? ;)

Thx
Oli

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 500
Credit: 474,948,762
RAC: 1,459,694
Level
Gln
Scientific publications
wat
Message 54070 - Posted: 25 Mar 2020 | 23:09:29 UTC - in response to Message 54068.

Thx for the reply and pointing me to the right article.

Seems the laptops will stay very cool... :(
Unless someone has a trick ?? ;)

Thx
Oli

You would have to purchase a gaming laptop with a discrete Nvidia gpu.

GPUGrid can only use Nvidia gpus. Not AMD or Intel.

Alex
Send message
Joined: 17 Feb 14
Posts: 1
Credit: 7,455,961
RAC: 40,124
Level
Ser
Scientific publications
wat
Message 54083 - Posted: 26 Mar 2020 | 16:56:09 UTC

Congratulations. I just bought a new pc recently, when I don't play games I can help a little.

popandbob
Send message
Joined: 18 Jul 07
Posts: 67
Credit: 25,576,692
RAC: 148,778
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54084 - Posted: 26 Mar 2020 | 17:38:10 UTC - in response to Message 54067.

I really think F@H needs to work in BOINC so that way if units are not being sent out you can do other work.

Those Stanford guys are too arrogant to use BOINC. I've been telling them that for years. It's hopeless.


They did for a while - they gave it up due to the extra headache and no real benefit to them.

That said though - a smart programmer could set it up quite easily as a third party.

Hotohori
Send message
Joined: 26 Mar 20
Posts: 2
Credit: 149,664
RAC: 18
Level

Scientific publications
wat
Message 54086 - Posted: 26 Mar 2020 | 18:45:11 UTC

I'm here because of the same reason than others here: F@H has too less WUs and i didn't want to waste PC time when i let my PC run 24/7 for Covid19 stuff.

I already use rosetta@home for CPU and Boinc itself since over 10 years (seti@home and WCG projects) and i must say i like it much more than F@H.

So i hope i could help with my RTX GPU a bit more here than on F@H.

But how can you see if a task is a Covid19 task here? The rosetta one have it in his name.

KHAN_VADER
Send message
Joined: 26 Mar 20
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 54087 - Posted: 26 Mar 2020 | 19:27:49 UTC - in response to Message 53999.

I found BOINC after Folding at home was having a work shortage due to the COVID-19 effort. Much better UI. Plenty of PC's laying around to put to use, just wish there was a tax write-off for all this energy I'm using on behalf of science!

Erich56
Send message
Joined: 1 Jan 15
Posts: 695
Credit: 3,280,654,574
RAC: 549,837
Level
Arg
Scientific publications
watwatwatwatwatwat
Message 54092 - Posted: 26 Mar 2020 | 20:35:32 UTC - in response to Message 54086.

But how can you see if a task is a Covid19 task here? The rosetta one have it in his name.

I don't think that GPUGRID has Covid19 tasks available at this point

Hotohori
Send message
Joined: 26 Mar 20
Posts: 2
Credit: 149,664
RAC: 18
Level

Scientific publications
wat
Message 54097 - Posted: 26 Mar 2020 | 21:56:18 UTC - in response to Message 54092.

But how can you see if a task is a Covid19 task here? The rosetta one have it in his name.

I don't think that GPUGRID has Covid19 tasks available at this point

Ok, but it is planned?

stefan
Send message
Joined: 27 Mar 20
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 54104 - Posted: 27 Mar 2020 | 4:41:37 UTC - in response to Message 54070.



GPUGrid can only use Nvidia gpus. Not AMD or Intel.


Is this true? Why can one check the AMD mark in preferences then?

AMD GPU would be cool!

Kind regards

stefan

Erich56
Send message
Joined: 1 Jan 15
Posts: 695
Credit: 3,280,654,574
RAC: 549,837
Level
Arg
Scientific publications
watwatwatwatwatwat
Message 54105 - Posted: 27 Mar 2020 | 6:49:59 UTC - in response to Message 54104.


GPUGrid can only use Nvidia gpus. Not AMD or Intel.

Is this true?

yes, that's the case

tullio
Send message
Joined: 8 May 18
Posts: 167
Credit: 47,470,854
RAC: 202,093
Level
Val
Scientific publications
wat
Message 54108 - Posted: 27 Mar 2020 | 9:54:57 UTC

Folding@home director Greg Bowman says they have reached 470 petaflops, while a pool of supercomputers including Summit and Sierra hava totalled 330 petaflops.
I simple don't believe Bowman

____________

tazzduke
Send message
Joined: 7 Aug 09
Posts: 2
Credit: 25,811,732
RAC: 5,153
Level
Val
Scientific publications
watwatwatwat
Message 54109 - Posted: 27 Mar 2020 | 10:54:22 UTC

Greetings All

Here is an article that does a good job on explaining.

https://www.tomshardware.com/news/folding-at-home-breaks-exaflop-barrier-fight-coronavirus-covid-19

Regards

Flo
Send message
Joined: 15 Apr 09
Posts: 1
Credit: 385,939
RAC: 0
Level

Scientific publications
watwatwat
Message 54110 - Posted: 27 Mar 2020 | 12:24:25 UTC

I just reactivated my account, I think I was 10 years inactive. I activated solely due to support Covid research with my computing power..

kain
Send message
Joined: 3 Sep 14
Posts: 150
Credit: 562,363,527
RAC: 163,666
Level
Lys
Scientific publications
watwatwatwatwatwatwatwat
Message 54111 - Posted: 27 Mar 2020 | 14:07:02 UTC - in response to Message 54110.

I just reactivated my account, I think I was 10 years inactive. I activated solely due to support Covid research with my computing power..


Sadly you won't be able to crunch with GTX 285, it's too old.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2184
Credit: 15,800,211,961
RAC: 795,161
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54114 - Posted: 27 Mar 2020 | 14:58:13 UTC - in response to Message 53999.
Last modified: 27 Mar 2020 | 15:00:06 UTC

If I am not missing anything, GPUGRID volunteers doubled the number of active hosts in the last month. It may have been either due to the closing of seti@home, or a consequence of folding at home’s covid effort.



(Credits: Boincstats)

Interestingly the total awarded credits per day hasn't changed much:

Aurum
Avatar
Send message
Joined: 12 Jul 17
Posts: 247
Credit: 9,673,488,208
RAC: 7,293,473
Level
Tyr
Scientific publications
wat
Message 54115 - Posted: 27 Mar 2020 | 15:00:49 UTC

That's cuz the heavy metal's been here all along.

tullio
Send message
Joined: 8 May 18
Posts: 167
Credit: 47,470,854
RAC: 202,093
Level
Val
Scientific publications
wat
Message 54116 - Posted: 27 Mar 2020 | 15:07:44 UTC
Last modified: 27 Mar 2020 | 15:09:15 UTC

Rosetta@home is only one of the fifty odd BOINC projects. All of them use about 30 petaflops. The COVID-19 supercomputer pool reaches 330 petaflops. To say that
Folding@home reaches 470 petaflops is simply ridiculous.
Tullio
____________

Aurum
Avatar
Send message
Joined: 12 Jul 17
Posts: 247
Credit: 9,673,488,208
RAC: 7,293,473
Level
Tyr
Scientific publications
wat
Message 54117 - Posted: 27 Mar 2020 | 15:13:10 UTC - in response to Message 54116.

The COVID-19 supercomputer pool reaches 330 petaflops. Tullio

What's this pool???

____________

tullio
Send message
Joined: 8 May 18
Posts: 167
Credit: 47,470,854
RAC: 202,093
Level
Val
Scientific publications
wat
Message 54122 - Posted: 27 Mar 2020 | 19:16:14 UTC - in response to Message 54117.

It is a consortium of US firms and scientific institutions incluudin the top 1 and 2 of the Top500 list. Summit (1) alone reaches 148 petaflops.
Tullio
____________

Profile ServicEnginIC
Avatar
Send message
Joined: 24 Sep 10
Posts: 195
Credit: 1,428,224,852
RAC: 836,727
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54124 - Posted: 27 Mar 2020 | 21:34:40 UTC - in response to Message 54114.
Last modified: 27 Mar 2020 | 21:35:45 UTC

At Mar 27th 2020 Retvari Zoltan wrote:

Interestingly the total awarded credits per day hasn't changed much:

-1) I think that a signicative number of new hosts derived from Seti@home closing are based on AMD, Intel, or old Nvidia GPUs, currently not viable at GPUGrid...

-2) It's curious to see the spike derived from this affair.

At Jan 29th 2020 Toni wrote:
Hello, I have a similar problem (or rather anomaly) - my credit got quadrupled today from ~70k to ~280k. I noticed that I finished a long GPU task, but why is this one particular task worth so much?...

It was actually a mistake in the MDADeq series. Will be fixed in the next series.

GabrielB
Send message
Joined: 16 Mar 20
Posts: 3
Credit: 10,420,273
RAC: 173,328
Level
Pro
Scientific publications
wat
Message 54130 - Posted: 28 Mar 2020 | 4:58:13 UTC - in response to Message 54026.

I'm here because F@H caused me to look at my BOINC processes and I saw I had underutilized GPU resources. It doesn't hurt that SETI is done for now.

I can't post elsewhere and the contact email address isn't responding. Can anyone help me figure out why all my tasks fail? closing in on 2 weeks of crunching and no support to make it fruitful. http://www.gpugrid.net/results.php?userid=554223

Yavanius
Avatar
Send message
Joined: 15 Jul 15
Posts: 2
Credit: 471,746
RAC: 4
Level

Scientific publications
wat
Message 54132 - Posted: 28 Mar 2020 | 7:40:14 UTC - in response to Message 54027.


I really think F@H needs to work in BOINC so that way if units are not being sent out you can do other work.


Some years they had a grad student working on that. Sadly, he moved on. They hardly have had the resources to update the current client. Dunno if they are getting more funds now.

With nVidia pointing out Folding@home, all the gamers went over there and their servers got overwhelmed (imagine your a restaurant and suddenly some famous reviewer says you gotta try them...it's not the restaurant's fault they don't have enough seating for all these people). They didn't have any advance notice and they can't just throw more server resources at it like the big corporations. It takes some time to get them and prep them.

You can get work, but there are times when the server just isn't able to get to uploading or downloading. Some folks at Rosetta said do Folding on the GPU and let the CPUs take up Rosetta so things are always crunching...

~Yav

Yavanius
Avatar
Send message
Joined: 15 Jul 15
Posts: 2
Credit: 471,746
RAC: 4
Level

Scientific publications
wat
Message 54133 - Posted: 28 Mar 2020 | 7:40:18 UTC - in response to Message 54027.
Last modified: 28 Mar 2020 | 7:41:01 UTC

Forum duplicated my message. O.o

rod4x4
Send message
Joined: 4 Aug 14
Posts: 158
Credit: 1,825,201,071
RAC: 977,598
Level
His
Scientific publications
watwatwatwatwatwatwat
Message 54134 - Posted: 28 Mar 2020 | 8:24:47 UTC - in response to Message 54130.
Last modified: 28 Mar 2020 | 8:32:32 UTC

I'm here because F@H caused me to look at my BOINC processes and I saw I had underutilized GPU resources. It doesn't hurt that SETI is done for now.

I can't post elsewhere and the contact email address isn't responding. Can anyone help me figure out why all my tasks fail? closing in on 2 weeks of crunching and no support to make it fruitful. http://www.gpugrid.net/results.php?userid=554223


Update your Nvidia driver. Minimum supported version is r418.39. Your driver is too old

see FAQ here:
http://www.gpugrid.net/forum_thread.php?id=5002

As you have an old driver, you were recieving jobs from a faulty batch (PABLO_UCB_NMR_KIX). This compounded your issues.

John Twohy
Send message
Joined: 26 Dec 08
Posts: 1
Credit: 2,777,417
RAC: 2,613
Level
Ala
Scientific publications
wat
Message 54144 - Posted: 29 Mar 2020 | 3:16:08 UTC - in response to Message 53999.

hey you guys, I see a CPU party coming up at Rosetta challenge if your interested. https://www.boincstats.com/stats/challenge/team/chat/1074

happer
Send message
Joined: 22 Nov 11
Posts: 1
Credit: 1,000,165
RAC: 58
Level
Ala
Scientific publications
wat
Message 54146 - Posted: 29 Mar 2020 | 16:07:04 UTC

Die COVID, DIE !!!

ProDigit
Send message
Joined: 13 Nov 19
Posts: 6
Credit: 48,670,830
RAC: 592,965
Level
Val
Scientific publications
wat
Message 54150 - Posted: 29 Mar 2020 | 22:43:37 UTC

I've connected quite a few ARM processor boards around that time, to see if an ARM-based server was viable.
They made up of about 50 quad core hosts, but they don't amount to a lot of processing power, due to it's power saving feats over raw processing power.

I've used them for testing purposes, but never really did a lot of tasks on them.
Give it another month and purge the deactivated units, and those with less than 10 completed tasks, and the number will more than likely reflect the same number it was before.

I have been running more GPUGrid on my RTX2080 Tis lately, due to a lot of projects running out of Nvidia tasks in the past few weeks.

codrin
Send message
Joined: 29 Mar 20
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 54152 - Posted: 30 Mar 2020 | 6:51:16 UTC

Hi all,

I am also here in order to contribute to biology research as much as possible, in this pandemic context.

Is GPUGRID involved in any COVID-19 research?

Thanks for running this project!
-Codrin

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2184
Credit: 15,800,211,961
RAC: 795,161
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54153 - Posted: 30 Mar 2020 | 14:56:48 UTC - in response to Message 54152.

Is GPUGRID involved in any COVID-19 research?
Not yet.

GabrielB
Send message
Joined: 16 Mar 20
Posts: 3
Credit: 10,420,273
RAC: 173,328
Level
Pro
Scientific publications
wat
Message 54154 - Posted: 30 Mar 2020 | 20:39:00 UTC - in response to Message 54134.

I'm here because F@H caused me to look at my BOINC processes and I saw I had underutilized GPU resources. It doesn't hurt that SETI is done for now.

I can't post elsewhere and the contact email address isn't responding. Can anyone help me figure out why all my tasks fail? closing in on 2 weeks of crunching and no support to make it fruitful. http://www.gpugrid.net/results.php?userid=554223


Update your Nvidia driver. Minimum supported version is r418.39. Your driver is too old

see FAQ here:
http://www.gpugrid.net/forum_thread.php?id=5002

As you have an old driver, you were recieving jobs from a faulty batch (PABLO_UCB_NMR_KIX). This compounded your issues.

Thank you! DLing now!

L
Send message
Joined: 22 Mar 14
Posts: 41
Credit: 429,660,405
RAC: 250,913
Level
Gln
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 54161 - Posted: 31 Mar 2020 | 9:09:46 UTC - in response to Message 54153.
Last modified: 31 Mar 2020 | 9:12:47 UTC

Is GPUGRID involved in any COVID-19 research?
Not yet.

Why?

Sanford
Send message
Joined: 23 Nov 09
Posts: 1
Credit: 343,117,845
RAC: 118,524
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 54185 - Posted: 1 Apr 2020 | 3:19:21 UTC - in response to Message 53999.

I've heard that Folding@Home has a ton of new clients, due to COVID-19 work. Does that somehow show up as new hosts for GPUGrid?

Erich56
Send message
Joined: 1 Jan 15
Posts: 695
Credit: 3,280,654,574
RAC: 549,837
Level
Arg
Scientific publications
watwatwatwatwatwat
Message 54187 - Posted: 1 Apr 2020 | 8:50:32 UTC - in response to Message 54185.

no - why should it?

Wrend
Send message
Joined: 9 Nov 12
Posts: 51
Credit: 522,101,722
RAC: 32,710
Level
Lys
Scientific publications
watwatwatwatwatwat
Message 54195 - Posted: 1 Apr 2020 | 17:13:03 UTC - in response to Message 54161.
Last modified: 1 Apr 2020 | 17:53:31 UTC

Is GPUGRID involved in any COVID-19 research?
Not yet.

Why?

Sorry, I've been busy playing Mario with my daughter who's home from school...

https://i.imgur.com/X4VfwXM.jpg

Also, I don't work on making these projects, just sometime crunch for them.

YMMV

...

But yeah, in all seriousness though, it looks like they're working on getting it ready still. I'm still crunching here and for Rosetta for the time being either way. It'll be nice to see COVID-19 related WUs from GPUGrid too, when they're ready.

Also, it looks likes some of the current WUs do have implications for potentially aiding in COVID-19 related research. See here for more info. http://www.gpugrid.net/forum_thread.php?id=5089&nowrap=true#54172
____________
My BOINC Cruncher, Minecraft Multiserver, Mobile Device Mainframe, and Home Entertainment System/Workstation: http://www.overclock.net/lists/display/view/id/4678036#

nucco
Send message
Joined: 28 Mar 20
Posts: 1
Credit: 6,278,065
RAC: 11,424
Level
Ser
Scientific publications
wat
Message 54242 - Posted: 5 Apr 2020 | 13:18:01 UTC - in response to Message 54002.
Last modified: 5 Apr 2020 | 13:21:25 UTC

I added my machine simply because of the COVID push. I didn't really care about SETI@home in the past, and thought it to be a poor reason for raising my electricity bill running my computer pretty much 24x7 at nearly max load.

If my computer doesn't die from overheating, then I imagine this will continue for as long as I don't feel alarmed about my electricity bills.

I've already experienced two thermal protection events and have ordered a new case fan -- I never had to worry about cooling in my typical usage in the past for this custom build full of careless mistakes (for example, I found out that my CPU cooler fan was installed backwards for the last 3 years after the first thermal protection cutout :) ).

Edit:
To be clear, I have always done some BOINC work, but only decided to add some GPU projects after I read about the Covid push (on Anandtech) and noticed that my GPU is pretty much always idle. I have Folding@Home installed now too.

CliffJumper
Send message
Joined: 5 Apr 20
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 54264 - Posted: 6 Apr 2020 | 16:08:49 UTC

I have a dumb question. If my PC does not have a compatible GPU, i assume I cannot participate in GPU grid?

I have the following : I know its old.

Mon Apr 6 15:45:48 2020 | | Starting BOINC client version 7.9.3 for x86_64-pc-linux-gnu
Mon Apr 6 15:45:48 2020 | | log flags: file_xfer, sched_ops, task
Mon Apr 6 15:45:48 2020 | | Libraries: libcurl/7.58.0 OpenSSL/1.1.1 zlib/1.2.11 libidn2/2.0.4 libpsl/0.19.1 (+libidn2/2.0.4) nghttp2/1.30.0 librtmp/2.3
Mon Apr 6 15:45:48 2020 | | Data directory: /var/lib/boinc-client
Mon Apr 6 15:45:48 2020 | | No usable GPUs found
Mon Apr 6 15:45:49 2020 | | [libc detection] gathered: 2.27, Ubuntu GLIBC 2.27-3ubuntu1
Mon Apr 6 15:45:49 2020 | | Host name: XXXXXXXXX
Mon Apr 6 15:45:49 2020 | | Processor: 2 GenuineIntel Intel(R) Core(TM)2 Duo CPU E8600 @ 3.33GHz [Family 6 Model 23 Stepping 10]
Mon Apr 6 15:45:49 2020 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good nopl cpuid aperfmperf pni dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 xsave lahf_lm pti tpr_shadow vnmi flexpriority dtherm
Mon Apr 6 15:45:49 2020 | | OS: Linux Ubuntu: Ubuntu 18.04.4 LTS [4.15.0-91-generic|libc 2.27 (Ubuntu GLIBC 2.27-3ubuntu1)]
Mon Apr 6 15:45:49 2020 | | Memory: 3.75 GB physical, 3.75 GB virtual
Mon Apr 6 15:45:49 2020 | | Disk: 228.23 GB total, 207.94 GB free
Mon Apr 6 15:45:49 2020 | | Local time is UTC +0 hours
Mon Apr 6 15:45:49 2020 | | Config: GUI RPCs allowed from:
Mon Apr 6 15:45:50 2020 | | Last benchmark was 18358 days 15:45:46 ago
Mon Apr 6 15:45:50 2020 | | No general preferences found - using defaults
Mon Apr 6 15:45:50 2020 | | Reading preferences override file
Mon Apr 6 15:45:50 2020 | | Preferences:
Mon Apr 6 15:45:50 2020 | | max memory usage when active: 1922.04 MB
Mon Apr 6 15:45:50 2020 | | max memory usage when idle: 3459.68 MB
Mon Apr 6 15:45:50 2020 | | max disk usage: 205.40 GB
Mon Apr 6 15:45:50 2020 | | don't use GPU while active
Mon Apr 6 15:45:50 2020 | | suspend work if non-BOINC CPU load exceeds 25%
Mon Apr 6 15:45:50 2020 | | (to change preferences, visit a project web site or select Preferences in the Manager)
Mon Apr 6 15:45:50 2020 | | Setting up project and slot directories
.
.
.

Mon Apr 6 15:54:06 2020 | | Running CPU benchmarks
Mon Apr 6 15:54:06 2020 | | Suspending computation - CPU benchmarks in progress
Mon Apr 6 15:54:38 2020 | | Benchmark results:
Mon Apr 6 15:54:38 2020 | | Number of CPUs: 2
Mon Apr 6 15:54:38 2020 | | 3356 floating point MIPS (Whetstone) per CPU
Mon Apr 6 15:54:38 2020 | | 84780 integer MIPS (Dhrystone) per CPU
Mon Apr 6 15:54:39 2020 | | Resuming computation

Keith Myers
Send message
Joined: 13 Dec 17
Posts: 500
Credit: 474,948,762
RAC: 1,459,694
Level
Gln
Scientific publications
wat
Message 54268 - Posted: 6 Apr 2020 | 22:40:50 UTC

I have a dumb question. If my PC does not have a compatible GPU, i assume I cannot participate in GPU grid?

Yes, if you don't have a relatively modern Nvidia card with sufficient VRAM, you cannot crunch for this project.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2184
Credit: 15,800,211,961
RAC: 795,161
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54269 - Posted: 6 Apr 2020 | 23:49:43 UTC - in response to Message 54264.

Mon Apr 6 15:45:48 2020 | | Starting BOINC client version 7.9.3 for x86_64-pc-linux-gnu
Mon Apr 6 15:45:48 2020 | | log flags: file_xfer, sched_ops, task
Mon Apr 6 15:45:48 2020 | | Libraries: libcurl/7.58.0 OpenSSL/1.1.1 zlib/1.2.11 libidn2/2.0.4 libpsl/0.19.1 (+libidn2/2.0.4) nghttp2/1.30.0 librtmp/2.3
Mon Apr 6 15:45:48 2020 | | Data directory: /var/lib/boinc-client
Mon Apr 6 15:45:48 2020 | | No usable GPUs found
Mon Apr 6 15:45:49 2020 | | [libc detection] gathered: 2.27, Ubuntu GLIBC 2.27-3ubuntu1
Mon Apr 6 15:45:49 2020 | | Host name: XXXXXXXXX
Mon Apr 6 15:45:49 2020 | | Processor: 2 GenuineIntel Intel(R) Core(TM)2 Duo CPU E8600 @ 3.33GHz [Family 6 Model 23 Stepping 10]
Mon Apr 6 15:45:49 2020 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good nopl cpuid aperfmperf pni dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 xsave lahf_lm pti tpr_shadow vnmi flexpriority dtherm
Mon Apr 6 15:45:49 2020 | | OS: Linux Ubuntu: Ubuntu 18.04.4 LTS [4.15.0-91-generic|libc 2.27 (Ubuntu GLIBC 2.27-3ubuntu1)]

The text in red says that you don't have any usable GPUs (NVidia, AMD, or Intel - of course a Core2Duo don't have integrated iGPU), or you don't have the appropriate drivers installed for it. If you would have a state of the art GPU, you still couldn't use it while this message appears in the first lines of BOINC startup log.

spelger
Send message
Joined: 4 Jul 18
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 54271 - Posted: 7 Apr 2020 | 1:09:59 UTC

wow, talk about a "chicken and an egg" situation... to post a new message i need credit. but i have no credit which is what i wanted to post about. this seemed just as good a thread to jump in with this question...


so, how does one actually get work units? i was attached to this project quite a long time ago but never got units so i detached. now that seti is/has ended i decided to re-attach. for the last few days still no work. as far as i can tell my log when starting boinc does not indicate that my gpu is insufficient.

glad there is rosetta and world community grid.

-scott

CliffJumper
Send message
Joined: 5 Apr 20
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 54272 - Posted: 7 Apr 2020 | 2:02:29 UTC - in response to Message 54268.

Thank you so for Now it is CPUs only.

Aurum
Avatar
Send message
Joined: 12 Jul 17
Posts: 247
Credit: 9,673,488,208
RAC: 7,293,473
Level
Tyr
Scientific publications
wat
Message 54275 - Posted: 7 Apr 2020 | 14:28:20 UTC

Spegler, Be sure you have ACEMD3 checked in your Preferences. Also you need an Nvidia GPU with current drivers.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2184
Credit: 15,800,211,961
RAC: 795,161
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 54276 - Posted: 7 Apr 2020 | 18:39:59 UTC - in response to Message 54271.
Last modified: 7 Apr 2020 | 18:42:43 UTC

as far as i can tell my log when starting boinc does not indicate that my gpu is insufficient.
Your laptop has a GT 550M, which is a Fermi architecture GPU. It has compute capability 2.1. This GPU is too old for the GPUGrid client.
BTW do not use Windows 7 anymore.

spRocket
Send message
Joined: 27 Mar 20
Posts: 2
Credit: 10,674,147
RAC: 144,973
Level
Pro
Scientific publications
wat
Message 54277 - Posted: 8 Apr 2020 | 15:19:47 UTC - in response to Message 54268.

Yes, if you don't have a relatively modern Nvidia card with sufficient VRAM, you cannot crunch for this project.


Yup... I found my old 9400 GT and popped it into one of my old systems, only to find that it wouldn't get tasks from GPUGRID. The GTX 960, on the other hand, is happily crunching away.

Then I looked again at nvidi-smi on the 9400, and sure enough, it says "Not Supported" under the computing task list. Oh well.

spelger
Send message
Joined: 4 Jul 18
Posts: 2
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 54279 - Posted: 9 Apr 2020 | 1:17:04 UTC

Aurum and Retvari Zoltan, thanks for your replies. i did have that option checked but if the laptop is too ancient then that's it for me. a new machine is not in the cards at this time. and i like my Win7 machine. heck, i kind of miss windows 2000.

artgia pen
Send message
Joined: 11 Apr 20
Posts: 1
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 54305 - Posted: 12 Apr 2020 | 0:20:55 UTC - in response to Message 53999.
Last modified: 12 Apr 2020 | 0:24:13 UTC

I would like to see in mainstream informations media advertisement of rosetta@home project and other similar initiatives, so the question is: how to tell to big decision-makers and to people the existence of all this stuff to cooperate all together against covid-19??!
____________

Post to thread

Message boards : News : Doubled number of hosts