Contribute your computing resources to fight COVID-19

poisson

https://foldingathome.org/2020/03/15/coronavirus-what-were-doing-and-how-you-can-help-in-simple-terms/

Available for Windows, Mac and Linux. I just set mine up and letting my GTX1070 and Ryzen5-2600X rip.

Only issue was heat lol. I had to remove side cover or else my CPU was cooking due to the extraction fans not keeping up.

EDIT: I created a LES team so if you are keen on having your contribution to science counted as a LES effort, this is LES team code: 240518. You enter it in the identity section in the configuration screen.

lentro

A bit like fluidstack (provider.fluidstack.io) except this one actually helps the world. In guess I’ll run this on my idling servers now

poisson

I think as a community, our guys have plenty of computing power and let's use this to try and strike back at the virus!

poisson

In case anyone is wonder, this is legitimate. The lab is based in Stanford, and I got to know of it from a MIT-trained chemistry PhD friend who knows these stuff.

seriesn

Thank you!

mobile

stupid question: is this also hugging network bandwidth? or it's resource intensive but only spent few MBs per hour.
if you ever run this, share stats please

poisson

@mobile said:
stupid question: is this also hugging network bandwidth? or it's resource intensive but only spent few MBs per hour.
if you ever run this, share stats please

No it doesn't hog bandwidth. It downloads a small amount of data, unpacks, runs the algorithm to do protein folding, and then uploads it back to the server. Here's a sample of the logs of what my CPU slot is currently doing.


03:34:31:WU02:FS02:Assigned to work server 128.252.203.9
03:34:31:WU02:FS02:Requesting new work unit for slot 02: READY cpu:4 from 128.252.203.9
03:34:31:WU02:FS02:Connecting to 128.252.203.9:8080
03:34:33:WU02:FS02:Downloading 2.16MiB
03:34:34:WU02:FS02:Download complete
03:34:34:WU02:FS02:Received Unit: id:02 state:DOWNLOAD error:NO_ERROR project:13833 run:0 clone:3044 gen:2 core:0xa7 unit:0x0000000380fccb095e6e55e8cbadc0c8
03:34:34:WU02:FS02:Starting
03:34:34:WU02:FS02:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/cores.foldingathome.org/v7/lin/64bit/avx/Core_a7.fah/FahCore_a7 -dir 02 -suffix 01 -version 705 -lifeline 2367 -checkpoint 15 -np 4
03:34:34:WU02:FS02:Started FahCore on PID 11194
03:34:34:WU02:FS02:Core PID:11198
03:34:34:WU02:FS02:FahCore 0xa7 started
03:34:35:WU02:FS02:0xa7:*********************** Log Started 2020-03-16T03:34:34Z ***********************
03:34:35:WU02:FS02:0xa7:************************** Gromacs Folding@home Core ***************************
03:34:35:WU02:FS02:0xa7: Type: 0xa7
03:34:35:WU02:FS02:0xa7: Core: Gromacs
03:34:35:WU02:FS02:0xa7: Args: -dir 02 -suffix 01 -version 705 -lifeline 11194 -checkpoint 15 -np
03:34:35:WU02:FS02:0xa7: 4
03:34:35:WU02:FS02:0xa7:************************************ CBang *************************************
03:34:35:WU02:FS02:0xa7: Date: Nov 5 2019
03:34:35:WU02:FS02:0xa7: Time: 06:06:57
03:34:35:WU02:FS02:0xa7: Revision: 46c96f1aa8419571d83f3e63f9c99a0d602f6da9
03:34:35:WU02:FS02:0xa7: Branch: master
03:34:35:WU02:FS02:0xa7: Compiler: GNU 8.3.0
03:34:35:WU02:FS02:0xa7: Options: -std=c++11 -O3 -funroll-loops -fno-pie -fPIC
03:34:35:WU02:FS02:0xa7: Platform: linux2 4.19.0-5-amd64
03:34:35:WU02:FS02:0xa7: Bits: 64
03:34:35:WU02:FS02:0xa7: Mode: Release
03:34:35:WU02:FS02:0xa7:************************************ System ************************************
03:34:35:WU02:FS02:0xa7: CPU: AMD Ryzen 5 2600X Six-Core Processor
03:34:35:WU02:FS02:0xa7: CPU ID: AuthenticAMD Family 23 Model 8 Stepping 2
03:34:35:WU02:FS02:0xa7: CPUs: 12
03:34:35:WU02:FS02:0xa7: Memory: 31.38GiB
03:34:35:WU02:FS02:0xa7:Free Memory: 24.02GiB
03:34:35:WU02:FS02:0xa7: Threads: POSIX_THREADS
03:34:35:WU02:FS02:0xa7: OS Version: 5.3
03:34:35:WU02:FS02:0xa7:Has Battery: false
03:34:35:WU02:FS02:0xa7: On Battery: false
03:34:35:WU02:FS02:0xa7: UTC Offset: 8
03:34:35:WU02:FS02:0xa7: PID: 11198
03:34:35:WU02:FS02:0xa7: CWD: /var/lib/fahclient/work
03:34:35:WU02:FS02:0xa7:******************************** Build - libFAH ********************************
03:34:35:WU02:FS02:0xa7: Version: 0.0.18
03:34:35:WU02:FS02:0xa7: Author: Joseph Coffland joseph@cauldrondevelopment.com
03:34:35:WU02:FS02:0xa7: Copyright: 2019 foldingathome.org
03:34:35:WU02:FS02:0xa7: Homepage: https://foldingathome.org/
03:34:35:WU02:FS02:0xa7: Date: Nov 5 2019
03:34:35:WU02:FS02:0xa7: Time: 06:13:26
03:34:35:WU02:FS02:0xa7: Revision: 490c9aa2957b725af319379424d5c5cb36efb656
03:34:35:WU02:FS02:0xa7: Branch: master
03:34:35:WU02:FS02:0xa7: Compiler: GNU 8.3.0
03:34:35:WU02:FS02:0xa7: Options: -std=c++11 -O3 -funroll-loops -fno-pie
03:34:35:WU02:FS02:0xa7: Platform: linux2 4.19.0-5-amd64
03:34:35:WU02:FS02:0xa7: Bits: 64
03:34:35:WU02:FS02:0xa7: Mode: Release
03:34:35:WU02:FS02:0xa7:************************************ Build *************************************
03:34:35:WU02:FS02:0xa7: SIMD: avx_256
03:34:35:WU02:FS02:0xa7:********************************************************************************
03:34:35:WU02:FS02:0xa7:Project: 13833 (Run 0, Clone 3044, Gen 2)
03:34:35:WU02:FS02:0xa7:Unit: 0x0000000380fccb095e6e55e8cbadc0c8
03:34:35:WU02:FS02:0xa7:Reading tar file core.xml
03:34:35:WU02:FS02:0xa7:Reading tar file frame2.tpr
03:34:35:WU02:FS02:0xa7:Digital signatures verified
03:34:35:WU02:FS02:0xa7:Calling: mdrun -s frame2.tpr -o frame2.trr -x frame2.xtc -cpt 15 -nt 4
03:34:35:WU02:FS02:0xa7:Steps: first=500000 total=250000
03:34:36:WU02:FS02:0xa7:Completed 1 out of 250000 steps (0%)
03:35:48:WU02:FS02:0xa7:Completed 2500 out of 250000 steps (1%)
03:37:01:WU02:FS02:0xa7:Completed 5000 out of 250000 steps (2%)
03:38:11:WU02:FS02:0xa7:Completed 7500 out of 250000 steps (3%)
03:39:21:WU02:FS02:0xa7:Completed 10000 out of 250000 steps (4%)
03:40:31:WU02:FS02:0xa7:Completed 12500 out of 250000 steps (5%)


poisson

By the way, if you decide to contribute and there is constantly an error using your CPU, it is very likely because the library that is needed doesn't support too many cores. I tried with a ten core and it failed. So far I have been able to get it working setting number of cores at 4.

For my GPU, it works with no problems, but you need an Nvidia GPU with CUDA installed.

EDIT: you can set up multiple slots each with the number of cores you are comfortable with contributing so that you can contribute all your idle CPU power.

mobile

@poisson said:
By the way, if you decide to contribute and there is constantly an error using your CPU, it is very likely because the library that is needed doesn't support too many cores. I tried with a ten core and it failed. So far I have been able to get it working setting number of cores at 4.

For my GPU, it works with no problems, but you need an Nvidia GPU with CUDA installed.

EDIT: you can set up multiple slots each with the number of cores you are comfortable with contributing so that you can contribute all your idle CPU power.

that's interesting, probably need to put it on container like podman or openshift to makes it run efficiently in lots of cores workstation.

source code is more interesting for me so i hopped to this page
https://foldingathome.org/support/faq/opensource/
https://github.com/FoldingAtHome

---

03:34:35:WU02:FS02:0xa7: SIMD: avx_256
i wonder if they just listed this AVX, if there's a NEON port i'd imagine it's also viable to run in arm devices.
like x264 for example (this isn't that scientific though lol)

it's just sad to see people buying or gifted with raspberry pi (and it's siblings) without knowing what to do with it
i think people will be happy seeing their paperweight Pi contribute something in science with low power cost
that tiny spark of dopamine probably worth it for average joe. man im so sleepy

yoursunny

@mobile said:

it's just sad to see people buying or gifted with raspberry pi (and it's siblings) without knowing what to do with it
i think people will be happy seeing their paperweight Pi contribute something in science with low power cost
that tiny spark of dopamine probably worth it for average joe. man im so sleepy

My BeagleBone Black has been running SETI@home for two years.
I have Raspberry Pi 3B and 4B. It overheats if I run SETI@home.

Folding@home download page doesn't seem to have ARMv7 or ARMv8 clients.
People needs to start writing everything in Go so it's portable to every platform.

havoc

Looks like folding@home is swamped by the interest...no more project available temporarily.

Alternatively check out BOINC / world computing grid - same concept

13:05:52:WU00:FS00:Connecting to 65.254.110.245:8080
13:05:52:WARNING:WU00:FS00:Failed to get assignment from '65.254.110.245:8080': No WUs available for this configuration
13:05:52:WU00:FS00:Connecting to 18.218.241.186:80
13:05:53:WARNING:WU00:FS00:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration
13:05:53:ERROR:WU00:FS00:Exception: Could not get an assignment

terrorgen

It requires python-gnome2 to be installed. Will that pull any X related packages?

I don't want any X related stuff installed on my cli only small vps...

Plus, my Debian 10 installation complains about not finding this package...

Andrei

Putting 4 GPUs towards this:

Uptime report + real time cpu/ram/network/disk usage:
http://hetrixtools.com/report/uptime/6dacf09a7cbcba0b12c28e23ba129e2a/

Go team LowEndSpirit!
Cheers.

Mr_Tom

How often is the team stats updated? Can't see my username there yet lol.

king8654

Def legit, was part of a folding team for OCN years ago, can get very competitive and lots of fun. Better in the winter to save on heating your office

poisson

Just grabbed the LES team stats! The term "wu" stands for work units or basically number of jobs done.


Tue Mar 17 01:52:15 GMT 2020
team teamname score wu
240518 LowEndSpirit 153869 30


I contributed 73443 points and 8 work units in 24 hours (mostly GPU).

poisson

@Andrei said:
Putting 4 GPUs towards this:

Uptime report + real time cpu/ram/network/disk usage:
http://hetrixtools.com/report/uptime/6dacf09a7cbcba0b12c28e23ba129e2a/

Go team LowEndSpirit!
Cheers.

Four Tesla T4s! My little GTX1070 is blown away

And you contributed almost as much as me in what, four hours?

Tue Mar 17 00:51:57 GMT 2020
name score wu team
HetrixTools 66501 13 240518


poisson

In case anyone is wondering whether donating your computing resources really work, here's a link to previous work by the same team on the Ebola virus which they used distributed computing to find a pocket in Ebola viral protein that is is a potential site for drugs and confirmed the findings using actual experiments: https://www.biorxiv.org/content/10.1101/2020.02.09.940510v1.abstract

Here's the abstract:

Many proteins are classified as ‘undruggable,’ especially those that engage in protein-protein and protein-nucleic acid interactions. Discovering ‘cryptic’ pockets that are absent in available structures but open due to protein dynamics could provide new druggable sites. Here, we integrate atomically-detailed simulations and biophysical experiments to search for cryptic pockets in viral protein 35 (VP35) from the highly lethal Ebola virus. VP35 plays multiple essential roles in Ebola’s replication cycle, including binding the viral RNA genome to block a host’s innate immunity. However, VP35 has so far proved undruggable. Using adaptive sampling simulations and allosteric network detection algorithms, we uncover a cryptic pocket that is allosterically coupled to VP35’s key RNA-binding interface. Experimental tests corroborate the predicted pocket and confirm that stabilizing the open form allosterically disrupts RNA binding. These results demonstrate simulations’ power to characterize hidden conformations and dynamics, uncovering cryptic pockets and allostery that present new therapeutic opportunities.

poisson

Update on LES contributions to fight Covid-19 with computational resources:

Team Statistics

Thu Mar 19 01:50:51 GMT 2020
team teamname score wu
240518 LowEndSpirit 3771115 123


Individual Statistics

Thu Mar 19 01:50:51 GMT 2020
name score wu team
HetrixTools 3329922 72 240518
json 268505 20 240518
aaronng82 137837 14 240518
darB 26101 4 240518
Mr_Tom 6860 9 240518
Laton 900 1 240518
NekkiBasara 590 2 240518
Xenos 400 1 240518


Thanks everyone for the effort! If you would like to join, go to the link in the original post!

poisson

Quick update on LES contributions to fight Covid-19 with computational resources again. Going strong!

Team Statistics

Thu Mar 19 01:50:51 GMT 2020
team teamname score wu
240518 LowEndSpirit 4445640 196

Individual Statistics

Thu Mar 19 01:50:51 GMT 2020
name score wu team
HetrixTools 3329922 72 240518
json 697060 38 240518
aaronng82 317973 29 240518
darB 58296 9 240518
Mr_Tom 30468 34 240518
NekkiBasara 7001 10 240518
Xenos 4020 3 240518
Laton 900 1 240518


FreeBSD

Intensive CPU usage on VPS may break the rules of provider. Is this CPU intensive?

seanho

Very. The assumption is that you'd run this on dedis, VDS with unlimited CPU usage per ToS, or owned hardware (e.g., homelab).

Don't run F@H on a LES NAT VPS, just ... don't

mikho

Leaving this here

gestiondbi

Keep computing guys! Want to get out of the house soon!

WSS

@seanho said:
Don't run F@H on a LES NAT VPS, just ... don't

If you can sigstop/sigcont it without issue, you technically can run it and be compliant with TOS. It's still very much a dick move, though.

chimichurri

Hey guys, I was about to start a new thread about Folding at Home, but luckily found this one first, so I'm bumping it instead!

It's not that difficult to setup, but to make your lives easier, here goes a quick start for Linux:

1. Ubuntu (suggested because of easier setup)
optional: install wget (in case you don't have it yet)
sudo apt install wget
get the package
wget https://download.foldingathome.org/releases/public/release/fahclient/debian-stable-64bit/v7.6/fahclient_7.6.13_amd64.deb
install the package (see setup notes below)
sudo dpkg -i fahclient_7.6.13_amd64.deb
check if the client is working (there should be some progress printed every couple of mins.)
tail -f /var/lib/fahclient/log.txt

Setup notes:
You will be prompted for a name and a team [2] - you can leave both defaults, optionally use the team LES code mentioned in the first post: 240518.
For the allowed load, if it's an idling instance, I'd choose "full" (or "light" or "medium" if they're not completely idle, or your provider does not allow full CPU usage)

2. CentOS (slightly more steps needed here, go for Ubuntu if you can)
read https://www.redhat.com/sysadmin/covid-folding-home - packages are slightly old, refer to https://foldingathome.org/alternative-downloads/ for the newest version

NOTE: maybe it's just that, to paraphrase a certain meme, chimichurri cannot into CentOS, but I had problems here...
Even though I launched a 8 core instance, top was showing only 300% / load 3 instead of 800% / load 8 on Ubuntu - so I ended up killing it in favor of Ubuntu

Last but not least, while I am probably in the minority here, because technically I do not have an idling virtual/dedicated server at the moment, I do happen to have some unused Azure/GCP credits which are about to expire soon. In case you are in a similar situation, here are my suggestions on how to use them effectively (=maximize cpu / $):

Azure
launch Ubuntu on a Standard F4 (4 vcpus, 8 GiB memory) instance in West-US2
(since only up to 4 total cores are allowed on the free tier, and this is the only CPU optimized family available on the free trial)

GCP
launch Ubuntu on custom E2 instances (max out CPU vs. RAM, e.g. 8CPU-4GB) in Oregon
(based on: https://cloud.google.com/compute/docs/benchmarks-linux & https://cloud.google.com/compute/vm-instance-pricing)
NOTE: you could get even more bang for buck with preemptible (short-lived discount priced) VMs, but you'd have to restart them every 24h

Happy folding and go team LES!!!

sonic

If you want to earn extra bucks, you can also try with https://arkfah.com
They pay crypto currency to folders up to 100$/week.

InceptionHosting

@sonic said:
If you want to earn extra bucks, you can also try with https://arkfah.com
They pay crypto currency to folders up to 100$/week.

I hope it goes without saying but I will say it anyway, don’t run any folding on a VPS, ever.

sonic

@AnthonySmith said:

@sonic said:
If you want to earn extra bucks, you can also try with https://arkfah.com
They pay crypto currency to folders up to 100$/week.

I hope it goes without saying but I will say it anyway, don’t run any folding on a VPS, ever.

I run from my PC with 3070 GPU and earn up to25$/week in cryptocurrency named ARK (ark.io).

chimichurri

@sonic said:
If you want to earn extra bucks, you can also try with https://arkfah.com
I run from my PC with 3070 GPU and earn up to25$/week in cryptocurrency named ARK (ark.io).

Thanks for the tip, but my credits won't last that long so I guess I'll pass..

Also, do you really earn that much? As in, including your electricity costs?
https://www.techradar.com/news/nvidia-geforce-rtx-3070 says the card itself was some 500 USD - hope you didn't buy it specifically for mining ^^'

Sorry for being so skeptical, but I remember mining bitcoin at home some 6 years ago, during a time when the whole crypto thing wasn't as hyped/widespread, and I imagine the market to be much more competitive now (although maybe it's easier because there are so many coins out there nowadays).
Back then, I bought a graphics card specifically for mining, and so for a while I was paying x times my usual power bill (although it wasn't money purely 'wasted' on mining, it was a source of heat during winter too ).
Unfortunately, the difficulty kept growing, leading to a quick end of this side income bonanza within a few months, which then led to me selling the GPU... ahhh, the good old times

sonic

@chimichurri said:

@sonic said:
If you want to earn extra bucks, you can also try with https://arkfah.com
I run from my PC with 3070 GPU and earn up to25$/week in cryptocurrency named ARK (ark.io).

Thanks for the tip, but my credits won't last that long so I guess I'll pass..

Also, do you really earn that much? As in, including your electricity costs?
https://www.techradar.com/news/nvidia-geforce-rtx-3070 says the card itself was some 500 USD - hope you didn't buy it specifically for mining ^^'

Sorry for being so skeptical, but I remember mining bitcoin at home some 6 years ago, during a time when the whole crypto thing wasn't as hyped/widespread, and I imagine the market to be much more competitive now (although maybe it's easier because there are so many coins out there nowadays).
Back then, I bought a graphics card specifically for mining, and so for a while I was paying x times my usual power bill (although it wasn't money purely 'wasted' on mining, it was a source of heat during winter too ).
Unfortunately, the difficulty kept growing, leading to a quick end of this side income bonanza within a few months, which then led to me selling the GPU... ahhh, the good old times

I run it for fun, because my works require online 24/7. This help me pay electricity costs